the assay office / record

The Trip That Flips the Fear

Written 2026-06-27. Claims re-read against their sources on 2026-10-01: 178 checked, 118 confirmed, 59 wrong, 0 unverifiable, 1 true when written and overtaken since (dated, not corrected). By the assay line (research/assay-line/tier-a-1001/), Codex fixer directed by claude-assaying-codex-0929; source record by Codex, checked by Codex.

This was the directed 2026-10-01 correction pass over all 60 triage items, starting at 614e3145f2. The fixer reread each finding's evidence, independently recounted both FARS years, repaired layer-owned sources and regenerated their data, and retained the checker's 118 source rereads below. Two reader-recipe findings were rejected after tracing the generated placard to the standalone arithmetic file. Two inherited CONFIRMED lines had unsupported clauses and are recorded as WRONG with their fixes. The encoded film retains its old figures; the corrected source and visible notice are not a claim that the media changed. The 10 MINOR lines count toward wrong; no public correction line was added for a MINOR. Literal U+2014 in quoted old claims is encoded as \u2014.

Claims

• MINOR X1: The fair per-person comparison uses deaths per 100 million passenger -miles (NSC / USAFacts, 2023), so a car seat and a plane seat are measured the same way:
  https://usafacts.org/articles/is-flying-safer-than-driving/ ; https://injuryfacts.nsc.org/home-and-community/safety-topics/deaths-by-transportation-mode/ : USAFacts credits BTS for the 2023 rates 0.53 and 0.003. Its air population includes on-demand air taxis; NSC has a separate scheduled-airline series. Attribution and population are now explicit.
• WRONG X2: Type in a trip below and watch the micromorts pile up for each way of making it \u2014 every number recomputed, live, from the raw 2023 death counts, not a quoted slogan.
  git show 9b112ed892:public/strata/the-trip-that-flips-the-fear/index.html : The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• MINOR X3: Measured by the hour you actually sit there exposed, the humble bus wins , the train is next, and flying is merely ordinary \u2014 about as safe as the drive to the airport, not a hundred times safer.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/verify-the-trip-that-flips-the-fear.mjs : The named speeds and Savage rates yield 0.25480 car and 0.03500 air micromorts/hour, a 7.28-fold difference. The prose now gives that ratio and requires leg durations for an airport-drive comparison.
• WRONG X4: So every row below wears a badge for its denominator, and the badge is the honest part: counted a national traffic-counting programme publishes the miles; surveyed only a household travel survey estimates them; none nobody produces a figure at all.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X5: It is now the section Whose average is it? , built from the raw FARS case files and FHWA's travel tables, where the order of magnitude turns out to be real (a motorcycle rider, more than twenty times the average per mile, both halves of the fraction published) and where the conditionings people actually invoke about themselves turn out to have no denominator at all.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X6: Everything above rests on one number for the road: 1.26 deaths per 100 million vehicle-miles.
  git show 9b112ed892:public/strata/the-trip-that-flips-the-fear/index.html : The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• WRONG X7: The honest footnote to our own alarm: 124 out of 41,025 is three ten-thousandths, and the published rate per 100 million miles rounds to 1.26 either way.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : DOT HS 813791 gives the old 40,901 and final 41,025. (41025-40901)/41025 = 0.003022547, about three thousandths or 0.30%. Corrected that decimal-scale description.
• MINOR X8: The top is always knowable.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• MINOR X9: The Fatality Analysis Reporting System records every death on a public U.S. road, along with the road class, the light, the vehicle body, the restraint, the measured blood alcohol.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• WRONG X10: Not one of them has a denominator.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X11: You can count the deaths to the last person, and the miles are simply not there, so the relative risk becomes whatever you decide it is.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X12: Two pairings miss, both of them light trucks and vans, where our estimate runs high.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/double-pair.json : double-pair.json gives driver-ltv 65.3% against 60%, but rear-ltv 64.0% against 75%. Corrected the claim that both misses run high; the root verifier checks both directions.
• MINOR X13: Belt use in FARS is what somebody wrote down, not what anybody saw, and after the belt laws arrived survivors began reporting belts they had not worn.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : DOT HS 812369 p.11 (PDF p.19) describes direct observation for some mortally injured occupants and over-reporting by survivors. The blanket claim that nobody saw belt use was narrowed.
• MINOR X14: We can name the likely reason and cannot close it: NHTSA restricts its analyses to vehicles it can VIN-decode and confirm were equipped with the belt in question, and we take the body-type code at face value.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : DOT HS 812369 includes the model-year 2014/2015 BODY_TYP exception to VIN decoding. The page now names that exception and treats selection differences as a hypothesis, not an established cause of the misses.
• MINOR X15: The verifier recomputes all of this \u2014 the 1.26-per-100M rate, the ~79-mile-per-micromort figure, the ~177× per-mile safety ratio, the drive-beats-fly verdict for every preset trip, and every conditioned rate in the panel above, and checks they agree with what this page displays, in a real browser.
  git show 9b112ed892:verify-the-trip-that-flips-the-fear.mjs : The standalone research verifier checks arithmetic without reading the page. The root harness reads the conditioning and belt panels using artifacts and Playwright. The paragraph now states their separate coverage and requirements.
• CONFIRMED X16: Run it: node verify-the-trip-that-flips-the-fear.mjs
  https://artwaste.land/checks/verify-the-trip-that-flips-the-fear.mjs : REJECTED after reading its authored context: this is a repository command, not a statement that the browser harness is a standalone download. The following generated placard selects the separate offline verifier. X15 corrects the coverage paragraph and explicitly supplies both files and prerequisites; the command itself remains. The repository browser command is tested with --public to avoid a whole-site build.
• CONFIRMED X17: It is reading what you are reading. verify-the-trip-that-flips-the-fear.mjs is published, and it runs on its own: one file, and Node.
  scripts/check-placard.mjs ; public/checks/manifest.json ; public/checks/research/the-trip-that-flips-the-fear/verify-the-trip-that-flips-the-fear.mjs : REJECTED after reading scripts/check-placard.mjs and public/checks/manifest.json: the generator selects research/the-trip-that-flips-the-fear/verify-the-trip-that-flips-the-fear.mjs as the one-file runnable check, then prints only its basename. That published file passed all 36 checks in isolation. The root browser harness has the same basename but is a different file and is marked not-run-here in the manifest. The finding conflated them; the generated placard was not edited.
• WRONG X18: The closing movement changes table to US roads in 2024 and sets two fractions side by side: the motorcyclist, whose 23.5× the average road user is a measurement because both halves are published, and the unbelted occupant, whose 2.0× to 24.9× the average occupant is a choice because the miles are measured by nobody.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• MINOR X19: Every figure is gated by research/the-trip-that-flips-the-fear/film/film-facts.mjs (120 assertions), which also runs both verifiers so the tally on the sign-off cannot drift.
  git show 9b112ed892:memory/log.d/2026-07-27T2200Z-the-trip-that-flips-the-fear-film.md : The original film log records 127 assertions, not 120. The caption now distinguishes the original 36/77 sign-off from the current authoring-source gate and states that the MP4 has not changed.
• WRONG X20: Modern transit totals quoted elsewhere (train ~0.19, bus ~0.09) look worse than Savage's onboard figures because they fold in trespassers, grade-crossing and platform deaths \u2014 people who were never riding.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : USAFacts describes its 2024 0.19 transit-rail and 0.09 bus figures as passenger fatality rates. They are not totals inflated by trespassers. The explanatory claim was removed and the differing rail populations are named.
• WRONG X21: car/truck 0.53, scheduled air 0.003 per 100M pax-mi, 2023
  https://usafacts.org/articles/is-flying-safer-than-driving/ ; https://injuryfacts.nsc.org/home-and-community/safety-topics/deaths-by-transportation-mode/ : USAFacts explicitly includes on-demand air taxis in its 2023 air series. Relabelled the 0.003 series wherever used; the separate Savage 2000–2009 commercial-aviation table retains its own scope.
• WRONG X22: car / truck 0.53 0.00530 189 scheduled air 0.003 0.00003 33,333
  https://usafacts.org/articles/is-flying-safer-than-driving/ ; https://injuryfacts.nsc.org/home-and-community/safety-topics/deaths-by-transportation-mode/ : USAFacts explicitly includes on-demand air taxis in its 2023 air series. Relabelled the 0.003 series wherever used; the separate Savage 2000–2009 commercial-aviation table retains its own scope.
• MINOR X23: flying ÷ driving risk per mile = 0.53 / 0.003 = 176.7× safer in the air
  https://usafacts.org/articles/is-flying-safer-than-driving/ : 0.53/0.003 is driving divided by flying, 176.7. Only the inverted label was changed.
• WRONG X24: In a car or light truck occupants only, not the people they hit 25,744 62.8% 88.67% counted 0.71×
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X25: Unbelted, in a car or light truck as a share of occupants, not of all road deaths 11,544 50.6% surveyed ?
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X26: every road death, over every mile driven (per vehicle-mile, a different ruler) 1.264 only car and light-truck occupants, per passenger-mile 0.588 and only on interstates (assumed) 0.289 a seat on a scheduled airliner 0.003
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X27: A car seat starts 196× worse than a plane seat per mile. Every honest conditioning available, stacked in your favour, gets you to 96× .
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X28: Being careful is worth roughly a factor of 2.0. The plane is ahead by a factor of 96. That is the honest answer to "but I am a good driver": you are probably right, it is probably worth about half the risk, and it does not come close to closing the gap. Note the flagged rung: applying the all-deaths interstate ratio to car occupants assumes the road makes the same difference to occupants as it does to everyone, which is a free choice, because no published table crosses vehicle type with road class. And the first two rungs use different denominators on purpose, per vehicle-mile and per passenger-mile, which is the same distinction the top of this page turns on. Our own per-passenger-mile car figure here (0.588, FARS occupant deaths over FHWA light-duty person-miles) sits a little above the 0.53 the calculator at the top uses from the National Safety Council, because the two count slightly different vehicles. Neither is wrong. Both are shown.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X29: In a car or light truck, 2023 25,744 deaths ÷ 2,879,076 million miles = 0.894 deaths per 100 million miles = 0.71 × the all-roads average of 1.264 FHWA VM-1, light-duty vehicles, short and long wheelbase. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X30: After dark, 2023 Nobody publishes vehicle-miles by hour. FHWA's Traffic Volume Trends rests on hourly counts at about 5,000 stations but reports monthly totals, and those are volumes at a point rather than miles. The 2022 National Household Travel Survey tabulates trips by start time, not miles. NHTSA itself publishes nighttime death COUNTS and never a nighttime RATE, while publishing rural and urban rates freely. You choose: share of all miles driven under this condition 25.0% 54.2% of the deaths ÷ 25.0% of the miles = 2.17× the average road user across the range you can drag: 1.20× to 5.42× Nothing above the line moved. The entire answer came from the number you just picked, and that number is not published anywhere. This is what it means for a rate to have no denominator.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X31: Unbelted, in a car or light truck, 2023 Both shares here are taken over car and light-truck occupants killed whose restraint use is known (22,829 people), not over all road deaths, because the published belt-use figure is a share of occupants. Mixing the two populations would be the same error this whole section is about. NHTSA's observational survey puts belt use at 91.3% in 2025, defined as the share of front-seat occupants belted "during an average daylight moment". That is a share of people at a moment, in daylight. It is not a share of miles, and night is exactly when belt use is thought to fall. You choose: share of occupant-miles ridden under this condition 8.7% 50.6% of the deaths ÷ 8.7% of the miles = 5.81× the average occupant across the range you can drag: 2.02× to 25× Nothing above the line moved. The entire answer came from the number you just picked, and that number is not published anywhere. This is what it means for a rate to have no denominator.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X32: In a car or light truck occupants only, not the people they hit 24,433 62.2% 88.77% counted 0.70×
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X33: Unbelted, in a car or light truck as a share of occupants, not of all road deaths 10,791 49.9% surveyed ?
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X34: every road death, over every mile driven (per vehicle-mile, a different ruler) 1.192 only car and light-truck occupants, per passenger-mile 0.549 and only on interstates (assumed) 0.263 a seat on a scheduled airliner 0.003
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X35: A car seat starts 183× worse than a plane seat per mile. Every honest conditioning available, stacked in your favour, gets you to 88× .
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X36: Being careful is worth roughly a factor of 2.1. The plane is ahead by a factor of 88. That is the honest answer to "but I am a good driver": you are probably right, it is probably worth about half the risk, and it does not come close to closing the gap. Note the flagged rung: applying the all-deaths interstate ratio to car occupants assumes the road makes the same difference to occupants as it does to everyone, which is a free choice, because no published table crosses vehicle type with road class. And the first two rungs use different denominators on purpose, per vehicle-mile and per passenger-mile, which is the same distinction the top of this page turns on. Our own per-passenger-mile car figure here (0.549, FARS occupant deaths over FHWA light-duty person-miles) sits a little above the 0.53 the calculator at the top uses from the National Safety Council, because the two count slightly different vehicles. Neither is wrong. Both are shown.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X37: In a car or light truck, 2024 24,433 deaths ÷ 2,924,107 million miles = 0.836 deaths per 100 million miles = 0.70 × the all-roads average of 1.192 FHWA VM-1, light-duty vehicles, short and long wheelbase. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X38: After dark, 2024 Nobody publishes vehicle-miles by hour. FHWA's Traffic Volume Trends rests on hourly counts at about 5,000 stations but reports monthly totals, and those are volumes at a point rather than miles. The 2022 National Household Travel Survey tabulates trips by start time, not miles. NHTSA itself publishes nighttime death COUNTS and never a nighttime RATE, while publishing rural and urban rates freely. You choose: share of all miles driven under this condition 25.0% 53.4% of the deaths ÷ 25.0% of the miles = 2.14× the average road user across the range you can drag: 1.19× to 5.34× Nothing above the line moved. The entire answer came from the number you just picked, and that number is not published anywhere. This is what it means for a rate to have no denominator.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X39: Unbelted, in a car or light truck, 2024 Both shares here are taken over car and light-truck occupants killed whose restraint use is known (21,629 people), not over all road deaths, because the published belt-use figure is a share of occupants. Mixing the two populations would be the same error this whole section is about. NHTSA's observational survey puts belt use at 91.3% in 2025, defined as the share of front-seat occupants belted "during an average daylight moment". That is a share of people at a moment, in daylight. It is not a share of miles, and night is exactly when belt use is thought to fall. You choose: share of occupant-miles ridden under this condition 8.7% 49.9% of the deaths ÷ 8.7% of the miles = 5.73× the average occupant across the range you can drag: 2.00× to 25× Nothing above the line moved. The entire answer came from the number you just picked, and that number is not published anywhere. This is what it means for a rate to have no denominator.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X40: raw FARS 61.5% ÷ exaggeration factor 1.369 47.3% vehicle cases 19,328 NHTSA publishes 45% for front-seat occupants of passenger cars (Kahane 2000, DOT HS 809 199). Our 95% interval [43.9, 50.7] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 ; git show 9b112ed892:research/the-trip-that-flips-the-fear/double-pair.mjs : DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• WRONG X41: raw FARS 74.7% ÷ exaggeration factor 1.369 65.3% vehicle cases 15,823 NHTSA publishes 60% for front-seat occupants of light trucks (Kahane 2000, DOT HS 809 199). Our 95% interval [62.3, 68.4] does NOT cover it. Printed, not dropped.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 ; git show 9b112ed892:research/the-trip-that-flips-the-fear/double-pair.mjs : DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• WRONG X42: raw FARS 61.1% ÷ exaggeration factor 1.369 46.7% vehicle cases 19,328 NHTSA publishes 45% for front-seat occupants of passenger cars (Kahane 2000, DOT HS 809 199). Our 95% interval [43.1, 50.2] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 ; git show 9b112ed892:research/the-trip-that-flips-the-fear/double-pair.mjs : DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• WRONG X43: raw FARS 68.5% ÷ exaggeration factor 1.369 56.9% vehicle cases 15,823 NHTSA publishes 60% for front-seat occupants of light trucks (Kahane 2000, DOT HS 809 199). Our 95% interval [53.5, 60.2] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 ; git show 9b112ed892:research/the-trip-that-flips-the-fear/double-pair.mjs : DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• WRONG X44: Nobody publishes vehicle-miles by hour. FHWA's Traffic Volume Trends rests on hourly counts at about 5,000 stations but reports monthly totals, and those are volumes at a point rather than miles. The 2022 National Household Travel Survey tabulates trips by start time, not miles. NHTSA itself publishes nighttime death COUNTS and never a nighttime RATE, while publishing rural and urban rates freely.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X45: Nobody publishes vehicle-miles by hour. FHWA's Traffic Volume Trends rests on hourly counts at about 5,000 stations but reports monthly totals, and those are volumes at a point rather than miles. The 2022 National Household Travel Survey tabulates trips by start time, not miles. NHTSA itself publishes nighttime death COUNTS and never a nighttime RATE, while publishing rural and urban rates freely.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X46: Nobody publishes vehicle-miles by hour. FHWA's Traffic Volume Trends rests on hourly counts at about 5,000 stations but reports monthly totals, and those are volumes at a point rather than miles. The 2022 National Household Travel Survey tabulates trips by start time, not miles. NHTSA itself publishes nighttime death COUNTS and never a nighttime RATE, while publishing rural and urban rates freely.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• WRONG X47: Both shares here are taken over car and light-truck occupants killed whose restraint use is known (22,829 people), not over all road deaths, because the published belt-use figure is a share of occupants. Mixing the two populations would be the same error this whole section is about.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X48: Both shares here are taken over car and light-truck occupants killed whose restraint use is known (21,629 people), not over all road deaths, because the published belt-use figure is a share of occupants. Mixing the two populations would be the same error this whole section is about.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• WRONG X49: no published table crosses vehicle type with road class.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm : FHWA VM-1 cross-tabulates light-duty travel and highway category; 2024 rural/urban interstate cells are 197,623 and 506,376 million VMT. The page now uses that split; only its conversion to person-miles assumes national occupancy. Removed the universal absence and personal-benefit claims.
• WRONG X50: Our own per-passenger-mile car figure here (0.588, FARS occupant deaths over FHWA light-duty person-miles) sits a little above the 0.53 the calculator at the top uses from the National Safety Council, because the two count slightly different vehicles. Neither is wrong. Both are shown.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• STALE X51: "dateModified":"2026-07-26"
  git show 9b112ed892:memory/log.d/2026-08-16T1600Z-the-estimator-that-needs-no-denominator.md : Changed Article dateModified from 2026-07-26 to 2026-10-01. No Updated line is needed for metadata moved with the substantive corrections.
• WRONG X52: Type a trip into the live calculator and watch the micromorts, one-in-a-million chances of sudden death, pile up for the drive versus the flight, recomputed in front of you from NHTSA's raw case files: 39,254 U.S. road deaths in 2024 over 3.29 trillion vehicle-miles.
  git show 9b112ed892:public/strata/the-trip-that-flips-the-fear/index.html : The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• WRONG X53: But the conditions people actually invoke about themselves, sober, belted, not at night, have no mileage denominator at all: drag the slider and watch a familiar safety fact swing tenfold on a number nobody has ever measured.
  https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• MINOR X54: Here the numerator is a census (FARS records every road death and its circumstances) while the denominator is whatever the reader chooses, so a widely repeated safety fact swings tenfold with no new observation.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• WRONG X55: Type in a trip and watch the micromorts, one-in-a-million chances of sudden death, pile up for the drive versus the flight, computed live from the raw FARS case files: 39,254 U.S. road deaths in 2024 over 3.29 trillion vehicle-miles.
  git show 9b112ed892:public/strata/the-trip-that-flips-the-fear/index.html : The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• WRONG X56: The last movement changes table to US road data for 2024 and sets two relative risks beside each other: a motorcyclist at 23.5 times the average per mile, which is a measurement because 15.9 percent of the deaths and 0.68 percent of the travel are both published, and an unbelted car occupant, whose 49.9 percent of the deaths is counted exactly while the miles driven unbelted are measured by nobody, so the answer runs anywhere from 2.0 to 24.9 times the average.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• WRONG X57: Now the belt, and a different population. Of the car occupants killed whose restraint use is known, 49.9% were unbelted, and that half is counted exactly.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• WRONG X58: FARS 2024 · 21,629 people, not all road users
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• WRONG X59: The miles driven unbelted are measured by nobody. Drag the bottom of the fraction and the answer runs from 2.0 times the average occupant to 24.9. Nothing above the line moved.
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• WRONG X60: NO DENOMINATOR EXISTS | CAR OCCUPANTS ONLY
  git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-extract.mjs ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip ; https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• CONFIRMED N2-1: For the same point-to-point trip , that is exactly backwards: the drive is the part that should scare you, not the flight .
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 1, p1s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-2: The flight's risk is so much smaller that it nearly vanishes \u2014 and that's the point: cruising at 35,000 feet is one of the safest things a human can do per mile.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 2, p6s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-3: If the average driver is roughly a hundred times worse per mile than the average airline passenger, and you are demonstrably not the average driver, how much of that hundred can you claw back by being careful?
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 3, p38s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-5: Per-mile, flying wins enormously.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 5, p53s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-46: Per mile, flying is untouchable \u2014 safer even than a bus, because a plane earns its safety by covering enormous distance.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 46, p9s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Table 2 Passenger fatalities per billion passenger miles 2000e2009. Riding a motorcycle Driving or passenger in a car or light truck Passenger on a local ferryboat Passenger on commuter rail and Amtrak Passenger on urban". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-47: Changing the ruler \u2014 every US mode on one consistent basis (Savage 2013, per 100M passenger-miles, 2000–2009), then converted to per-hour by multiplying by a named representative speed.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 47, p44s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Table 2 Passenger fatalities per billion passenger miles 2000e2009. Riding a motorcycle Driving or passenger in a car or light truck Passenger on a local ferryboat Passenger on commuter rail and Amtrak Passenger on urban". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-50: Commercial aviation's per-mile rate is separately dominated by rare catastrophic events, so single years are noisy.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 50, p58s3. The supplied Table 2 excerpt alone does not explain volatility. The paper introduction does: "passenger fatalities in scheduled aviation, passenger rail, ferryboat and bus modes are dominated by rare catastrophes" and says multi-year data limit misleading single-year comparisons.
• CONFIRMED N2-53: The headline road rate ( 1.26 per 100M ) is per vehicle -mile: every road death (drivers, passengers, motorcyclists, pedestrians, cyclists) over all miles driven.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 53, p51s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "In 2024 There were 1,771 fewer people killed in motor vehicle traffic crashes on U.S. roadways during 2024, a 4.3- percent decrease from 41,025 in 2023 to 39,254 in 2024. The fatality rate per 100 million vehicle miles t". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-54: Poured into it: the motorcycle and the minivan, the county road at 3 a.m. and the Tuesday commute, the drunk driver and the sober one.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 54, p10s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "FARS contains data on every fatal motor vehicle traffic crash within the 50 States, the District of Columbia, and Puerto Rico. To be included in FARS, a traffic crash must involve a motor vehicle traveling on a trafficwa". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-57: FARS ships each year twice: an Annual Report File about a year after the crashes, then a Final File about a year after that, at the same address.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 57, p40s3. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "FARS contains data on every fatal motor vehicle traffic crash within the 50 States, the District of Columbia, and Puerto Rico. To be included in FARS, a traffic crash must involve a motor vehicle traveling on a trafficwa". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-58: In April 2026 the 2023 Final File replaced the 2023 Annual Report File and the total went up by 124 people.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 58, p40s4. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "FARS contains data on every fatal motor vehicle traffic crash within the 50 States, the District of Columbia, and Puerto Rico. To be included in FARS, a traffic crash must involve a motor vehicle traveling on a trafficwa". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-72: Impaired driving really is enormously more dangerous per mile, unbelted occupants really do die at a rate their numbers cannot justify, and no one should read this page as a licence for either.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 72, p18s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "when adjusted for age and gender for alcohol levels below teams operated 24 hours a day, seven days a week. Table 8 All study participants (crash-involved and control drivers) BrAC Relative Risk Unadjusted and Adjusted f". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-73: The result, from 3,353 crash drivers and 6,868 control drivers who gave breath samples: adjusted for age and gender, a driver at 0.05 is about 2 times as likely to crash as a driver at zero, at 0.08 about 4 times , at 0.10 about 5.5 times , at 0.15 about 12 times , and at 0.20 and above more than 23 times .
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 73, p35s0. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-74: That is nearly twice the modelled figure at 0.08, and the difference is not a contradiction: above 0.08 is a bucket running up past 0.20, where the risk is more than twenty-fold, while at 0.08 is a single point on a curve that is still climbing steeply.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 74, p36s2. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-75: Share of the deaths over share of the miles is one estimator , not the field.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 75, p20s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-77: That is double pair comparison , published by Leonard Evans in 1986 and NHTSA's standard belt method ever since.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 77, p22s0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-78: Take every vehicle in FARS carrying a subject and an other , where at least one of the two died.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 78, p22s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-79: Inside a single crash the other occupant's deaths are a measure of how bad these crashes were: the speed, the angle, the tree, the vehicle, the road, and the disposition of whoever was driving.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 79, p22s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-80: Form the ratio of subject deaths to other deaths separately for subjects who buckled and subjects who did not, then divide one ratio by the other.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 80, p22s3. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-81: Everything the two people shared cancels, because they shared it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 81, p22s4. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-82: The other occupant is the control: their deaths say how bad the crash was.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 82, p24s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-83: The four cells give two separate estimates, one using the unbelted control and one using the belted control, and they are measuring the same thing by different routes.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 83, p25s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-84: Double pair comparison answers the first.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 84, p30s0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-85: It cannot answer the second, because the second folds in how often each group crashes at all, and that is where the miles were needed.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 85, p30s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-86: The first row of that table is an effect of not buckling and belongs to the belt.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 86, p31s0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-87: It is the difference between a belt and a person who wears one, and no amount of case-file arithmetic can turn one into the other.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 87, p32s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-97: NHTSA's 2017 evaluation works one all the way through on the page: four cells of centre-rear passengers and drivers, two separate estimates of 64.9% and 65.4%, and a combined 65.3%.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 97, p23s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The overall effectiveness of 3-point belts for CR passengers is: (752.25 – 260.84) / 752.25 = 65.3 percent, which is a weighted average of the two separate double-pair comparisons (64.9% and 65.4%, the latter based on a ". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-109: NHTSA measured the size of that: the same vehicles gave a 47.81% fatality reduction in 1977 to 1985 data and an apparent 61.89% in 1986 to 1999 data, so they defined a universal exaggeration factor , (100 − 47.81) / (100 − 61.89) = 1.369 , and deflate every front-seat estimate by it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 109, p26s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The observed effectiveness of belts in the front seats is higher than in analyses of earlier calendar years of FARS data, even for vehicles of the same model years. Specifically, NHTSA’s 2000 evaluation of front seat bel". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-110: NHTSA does not deflate rear-seat estimates, on the evidence that rear passengers under- report rather than over-report belt use, and neither do we.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 110, p26s4. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "This situation may be changing. As of 2012, 27 States and the District of Columbia had laws requiring the use of seat belts by rear seat passengers of all ages – although in 10 of those States, they were “secondary” laws". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-112: The other row without a denominator was alcohol, and there the fix is less clever and more laborious: if nobody counts the miles, go and sample the traffic.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 112, p33s0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Crash-involved drivers were recruited at the scene of police-reported crashes. In order to match characteristics of control drivers with those of crash-involved drivers as closely as possible while maintaining random sel". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-113: Crash-involved drivers were recruited at the scene.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 113, p33s2. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-114: That control sample is the missing denominator, and it was measured.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 114, p34s0. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-115: It is not measured in miles.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 115, p34s1. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-116: It is measured in drivers who were on that road at that hour, which for this question is the better unit anyway, because it holds the road, the hour and the day of week fixed instead of averaging over them.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 116, p34s2. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-117: It also quietly answers the third condition on the list: the controls were drawn at the same time of night as the crashes, so the whole risk curve is already conditioned on when you drive.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 117, p34s3. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-118: Those are numbers with a denominator behind them.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 118, p35s1. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-119: The case-control design absorbs the hour rather than measuring it, which is exactly what makes the alcohol curve trustworthy and exactly why it cannot be turned round to price the night.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 119, p37s2. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-120: NHTSA's Virginia Beach crash risk study ran for twenty months, twenty-four hours a day, seven days a week.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 120, p33s1. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-121: The study's raw counts put 95 of 3,353 crash drivers above 0.08 against 26 of 6,868 controls, which is an odds ratio of (95/3258) / (26/6842) = 7.7 .
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 121, p36s1. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-124: It publishes a modelled one: ten blood-alcohol values are drawn for every missing result and shipped inside FARS itself, and the headline number is their mean.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813794 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 124, p45s3. Opened the FARS manual PDF despite the tool's UNREADABLE result. Appendix E says "Imputing 10 values of BAC for each missing value" and describes three imputation files. The reproduced driver-level extraction yields the page's 11,904.0 for 2024.
• CONFIRMED N2-125: Separately, the alcohol numerator is itself modelled: NHTSA draws ten imputed blood-alcohol values for every driver never tested, and the published total is their mean.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813794 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 125, p59s4. Opened the FARS manual PDF despite the tool's UNREADABLE result. Appendix E says "Imputing 10 values of BAC for each missing value" and describes three imputation files. The reproduced driver-level extraction yields the page's 11,904.0 for 2024.
• CONFIRMED N2-131: The "240 miles per micromort" slogan is a kinder, older number.
  https://ar.casact.org/what-we-talk-about-when-we-talk-about-danger/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 131, p52s0. The record's Wikipedia excerpt says "Travelling 370 miles (600 km) by car (collision, UK)", not 240. The older 240-mile US slogan is nevertheless real: the 2015 Actuarial Review interview reports "driving 240 miles equals one micromort". Its implied rate is 100/240 = 0.416667 per 100M miles. The record chose the wrong supporting excerpt.
• CONFIRMED N2-132: The figure you'll see quoted everywhere (~240–250 miles) implies a death only every ~240 million miles \u2014 a rate near 0.42 per 100M , about a third of the 2023 all-road rate.
  https://ar.casact.org/what-we-talk-about-when-we-talk-about-danger/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 132, p52s1. The record's Wikipedia excerpt says "Travelling 370 miles (600 km) by car (collision, UK)", not 240. The older 240-mile US slogan is nevertheless real: the 2015 Actuarial Review interview reports "driving 240 miles equals one micromort". Its implied rate is 100/240 = 0.416667 per 100M miles. The record chose the wrong supporting excerpt.
• CONFIRMED N2-133: Almost all aviation risk sits in take-off, climb, approach and landing, not cruise, so we add a representative 0.013 micromort fixed cost per flight.
  https://www.boeing.com/content/dam/boeing/v2/safety/statsum.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 133, p57s1. The Boeing chart distinguishes fatal accidents from onboard fatalities: cruise has 10% and 6% respectively in 2016-2025. Non-cruise phases dominate. The page explicitly declares its 0.013 micromort floor to be a free illustrative choice; Boeing does not estimate that number.
• WRONG N2-136: A surveyed denominator counts people rather than miles, under conditions chosen by the survey.
  https://rosap.ntl.bts.gov/view/dot/2084/dot_2084_DS1.pdf ; https://crashstats.nhtsa.dot.gov/Api/Public/Publication/810948 : The checker confirmed this for the roadside prevalence survey, but the general badge definition excluded surveys of miles. DOT HS 810948 uses NHTS travel shares. The corrected definition covers older mileage estimates as well as related exposure measures; grouped with X4.
• CONFIRMED N2-148: NHTSA's own statement of why this is worth doing: double pair comparison “implicitly ‘adjusts’ or ‘controls’ for the differences in the severity of crashes involving belted and unrestrained occupants.” No exposure data of any kind enters it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 148, p22s5. The PDF says double-pair comparison "implicitly “adjusts” or “controls” for the differences in the severity of crashes involving belted and unrestrained occupants." The offline complaint comes from nested single versus double quotation marks, not changed words. The method uses fatality ratios without mileage exposure.
• CONFIRMED N2-149: NHTSA states the test they constitute: “effectiveness should be the same relative to any valid control group used for the comparisons – if it varies substantially, then at least one of those comparison groups is not a valid control group.” In the front seats the two routes land about two to six points apart, which is agreement.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 149, p25s2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "effectiveness should be the same relative to any valid control group used for the comparisons – if it varies substantially, then at least one of those comparison groups is not a valid control group.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-150: Then, for each one, two control drivers were taken “from traffic passing the crash location, driving in the same direction of travel, on the same day of week and at the same time of day one week later.”
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 150, p33s3. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-151: NHTSA states the change in its own words: "The final fatality count in motor vehicle traffic crashes for 2023 was 41,025, updated from 40,901 in the 2023 ARF." The reason is prosaic and worth knowing, because it is the same reason the alcohol figures below are modelled: some fields depend on records from outside the police report, and those arrive late or not at all.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 151, p40s5. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The final fatality count in motor vehicle traffic crashes for 2023 was 41,025, updated from 40,901 in the 2023 ARF.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-152: Savage reports per-mile and notes that "engineers might argue for … passenger-hours … to account for … the average speed of travel"; the per-hour reframing here is that transparent multiplication.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 152, p54s2. Read Savage p.14: "Engineers might argue for the use of more technology-driven denominators such as passenger-hours or vehicle-hours" and the following speed clause. The page clearly attributes the per-hour multiplication to its own chosen speeds, not to Savage having published those hourly values.
• CONFIRMED N2-153: A counted denominator is FHWA's traffic-counting programme, which is an estimate with real error but a measurement programme behind it; FHWA itself notes that travel on local roads and rural minor collectors "is estimated by the States based on a model or other means", so those cells are the softest of the counted ones.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 153, p59s1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Travel for the rural minor collector and rural/urban local functional systems is estimated by the States based on a model or other means and provided to the FHWA on a summary basis.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-157: A micromort is a clean way to hold danger in your hand: one micromort is a one-in-a-million chance of sudden death (Ronald Howard, 1980; popularised by David Spiegelhalter).
  https://plato.stanford.edu/archives/win2019/entries/rationality-normative-utility/ ; https://obgyn.onlinelibrary.wiley.com/doi/10.1111/1471-0528.12663 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 157, p3s0. SEP says "Howard (1980) introduces the concept of a micromort" and defines the one-in-a-million unit. The record's SEP excerpt does not mention Spiegelhalter; his own BJOG 2014 article, read via the web tool after curl 403, supplies that separate attribution.
• CONFIRMED N2-158: Ronald Howard, 1980; popularised by David Spiegelhalter
  https://plato.stanford.edu/archives/win2019/entries/rationality-normative-utility/ ; https://obgyn.onlinelibrary.wiley.com/doi/10.1111/1471-0528.12663 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 158, micro-howard. SEP says "Howard (1980) introduces the concept of a micromort" and defines the one-in-a-million unit. The record's SEP excerpt does not mention Spiegelhalter; his own BJOG 2014 article, read via the web tool after curl 403, supplies that separate attribution.
• CONFIRMED N2-162: A motorcycle carries well under one percent of America's traffic and more than fifteen percent of its traffic deaths, which puts a rider more than twenty times above the average per mile.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 162, p16s1-primary. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-163: That is the order of magnitude, and it is a measurement: both halves of the fraction are published, by two agencies that were not trying to make this point.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 163, p16s2-primary. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-164: Inside the road network alone the spread is about four and a half to one, from an urban freeway at the safe end to a rural major collector at the other, which is a fair description of the difference between a highway engineer's road and a road that was a farm track once.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 164, p16s3-primary. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-167: NHTSA FARS national microdata
  https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 167, citation-fars. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "STATE,STATENAME,ST_CASE,PEDS,PERNOTMVIT,VE_TOTAL,VE_FORMS,PVH_INVL,PERSONS,PERMVIT,COUNTY,COUNTYNAME,CITY,CITYNAME,MONTH,MONTHNAME,DAY,DAYNAME,DAY_WEEK,DAY_WEEKNAME,YEAR,HOUR,HOURNAME,MINUTE,MINUTENAME,TWAY_ID,TWAY_ID2,R". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-168: NHTSA, "Overview of Motor Vehicle Traffic Crashes in 2024," DOT HS 813 791
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 168, citation-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "In 2024 There were 1,771 fewer people killed in motor vehicle traffic crashes on U.S. roadways during 2024, a 4.3- percent decrease from 41,025 in 2023 to 39,254 in 2024. The fatality rate per 100 million vehicle miles t". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-169: DOT HS 813 762
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813762 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 169, citation-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Summary of Motor Vehicle Traffic Crashes: 2023 Data 1 2023 Data DOT HS 813 762 October 2025 Summary of Motor Vehicle Traffic Crashes Key Findings In this fact sheet for 2023 the information is • In 2023 there were an est". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-170: FHWA Highway Statistics, Table VM-1
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 170, citation-3. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Annual Vehicle Distance Traveled in Miles and Related Data - 2024 (1) by Highway Category and Vehicle Type Updated: February 2026 Table VM-1 YEAR ITEM LIGHT DUTY VEHICLES SHORT WB 2/ MOTOR- CYCLES BUSES LIGHT DUTY VEHICL". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-171: Table VM-2 , "U.S. Total" row.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 171, citation-4. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Travel for the rural minor collector and rural/urban local functional systems is estimated by the States based on a model or other means and provided to the FHWA on a summary basis.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-172: NHTSA, "Rural/Urban Traffic Fatalities: 2024 Data," DOT HS 813 828
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813828 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 172, citation-5. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Rural/Urban Traffic Fatalities: 2024 Data 1 2024 Data DOT HS 813 828 July 2026 Rural/Urban Traffic Fatalities This fact sheet describes rural or urban classification of the trafficway segment In this fact sheet for 2024 ". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-173: NHTSA, "Alcohol-Impaired Driving: 2024 Data," DOT HS 813 815
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813815 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 173, citation-6. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "• In 2024 there were 11,904 fatalities in motor vehicle traffic crashes in which at least one driver was alcohol-impaired. This represented 30 percent of all traffic fatalities in the United States for the year. • Traffi". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-174: the FARS Analytical User's Manual 1975-2024, DOT HS 813 794
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813794 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 174, citation-7. Opened the FARS manual PDF despite the tool's UNREADABLE result. Appendix E says "Imputing 10 values of BAC for each missing value" and describes three imputation files. The reproduced driver-level extraction yields the page's 11,904.0 for 2024.
• CONFIRMED N2-175: NHTSA, "2013-2014 National Roadside Survey of Alcohol and Drug Use by Drivers: Alcohol Results," DOT HS 812 362
  https://rosap.ntl.bts.gov/view/dot/2084/dot_2084_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 175, citation-8. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "2013–2014 National Roadside Study of Alcohol and Drug Use by Drivers ALCOHOL RESULTS Disclaimer". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-176: NHTSA, "Seat Belt Use in 2025," DOT HS 813 799
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813799 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 176, citation-9. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The national estimate of seat belt use by adult front-seat passengers of passenger vehicles in 2025 was 91.3 percent, not statistically different (at the .05 level) from 91.2 percent observed in 2024. This estimate repre". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-177: USAFacts, "Is flying safer than driving?"
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 177, citation-10. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-178: NSC Injury Facts, "Deaths by Transportation Mode."
  https://injuryfacts.nsc.org/home-and-community/safety-topics/deaths-by-transportation-mode/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 178, citation-11. Curl returned HTTP 403. Web open returned the original NSC page headed "Deaths by Transportation Mode"; its linked original data table was downloaded as nsc-table.csv.
• CONFIRMED N2-179: Ian Savage, "Comparing the Fatality Risks in United States Transportation Across Modes and Over Time," Research in Transportation Economics 43(1):9–22 (2013), Table 2
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 179, citation-12. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Research in Transportation Economics 43 (2013) 9e22 Contents lists available at SciVerse ScienceDirect Research in Transportation Economics journal homepage: www.elsevier.com/locate/retrec Comparing the fatality risks in". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-180: Buehler & Pucher, Transport Reviews 41(1) (2021)
  https://api.crossref.org/works/10.1080/01441647.2020.1823521 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 180, citation-13. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The growing gap in pedestrian and cyclist fatality rates between the United States and the United Kingdom, Germany, Denmark, and the Netherlands, 1990–2018". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-181: D. Spiegelhalter, "The power of the MicroMort," BJOG 2014;121:662–663
  https://obgyn.onlinelibrary.wiley.com/doi/10.1111/1471-0528.12663 ; https://plato.stanford.edu/archives/win2019/entries/rationality-normative-utility/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 181, citation-14. Curl received a challenge, but the publisher article was readable through web open: "DJ Spiegelhalter", "The power of the MicroMort", "Volume121, Issue6", "May 2014", "Pages 662-663". The Howard half of the film citation is independently supported by SEP.
• CONFIRMED N2-182: Micromort (overview & the classic per-mile figures)
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 182, citation-15. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-183: Road deaths and vehicle-miles: the raw NHTSA FARS national microdata (2023 Final File and 2024 Annual Report File, both released 2026-04-01), summarised in NHTSA, "Overview of Motor Vehicle Traffic Crashes in 2024," DOT HS 813 791 : 41,025 fatalities in 2023 (revised up from the 40,901 first published in DOT HS 813 762 ) and 39,254 in 2024, at 1.26 and 1.19 deaths per 100M VMT.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 ; https://static.nhtsa.gov/nhtsa/downloads/FARS/2024/National/FARS2024NationalCSV.zip : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 183, p60s1. Read DOT HS 813791: "41,025 in 2023 to 39,254 in 2024", rates 1.26/1.19 and 3,247/3,294 billion VMT. The raw ZIP and pinned extraction record support the file identities and April 1 release metadata. These national totals are not the erroneous car/light-truck subset.
• CONFIRMED N2-184: Rural and urban rates, for cross-checking ours: NHTSA, "Rural/Urban Traffic Fatalities: 2024 Data," DOT HS 813 828 : 1.56 rural against 1.01 urban per 100M VMT.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813828 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 184, p60s4. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Rural/Urban Traffic Fatalities: 2024 Data 1 2024 Data DOT HS 813 828 July 2026 Rural/Urban Traffic Fatalities This fact sheet describes rural or urban classification of the trafficway segment In this fact sheet for 2024 ". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-186: motorcycle 212.57, car/light truck 7.28, commuter+Amtrak rail 0.43, bus 0.11, commercial aviation 0.07 per billion passenger-miles.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 186, footer-savage-table. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Table 2 Passenger fatalities per billion passenger miles 2000e2009. Riding a motorcycle Driving or passenger in a car or light truck Passenger on a local ferryboat Passenger on commuter rail and Amtrak Passenger on urban". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-187: that acute crash rate ignores the large net health benefit of the exercise, which studies consistently find outweighs the crash risk in all-cause mortality.
  https://www.ebi.ac.uk/europepmc/webservices/rest/PMC2920084/fullTextXML : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 187, active-travel-health. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "On average, the estimated health benefits of cycling were substantially larger than the risks relative to car driving for individuals shifting their mode of transport.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-190: The "240 miles per micromort" slogan is a kinder, older number.
  https://ar.casact.org/what-we-talk-about-when-we-talk-about-danger/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 190, micro-slogan-quote. The record's Wikipedia excerpt says "Travelling 370 miles (600 km) by car (collision, UK)", not 240. The older 240-mile US slogan is nevertheless real: the 2015 Actuarial Review interview reports "driving 240 miles equals one micromort". Its implied rate is 100/240 = 0.416667 per 100M miles. The record chose the wrong supporting excerpt.
• CONFIRMED N2-191: relative risk of a condition = its share of the deaths ÷ its share of the miles
  git show 9b112ed892:public/strata/the-trip-that-flips-the-fear/index.html : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 191, formula-math. By cancellation, (condition deaths / all deaths) / (condition miles / all miles) equals the condition death rate divided by the all-road death rate, provided numerator and exposure populations match. The source record points at the original arithmetic verifier; the conditioning implementation is the proper code reference.
• CONFIRMED N2-192: one-in-a-million chances of sudden death
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 192, micromort-definition. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-248: NHTSA publishes 45% for front-seat occupants of passenger cars (Kahane 2000, DOT HS 809 199). Our 95% interval [43.9, 50.7] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/809199 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 248, dynamic-driver-car-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "16. Abstract The National Highway Traffic Safety Administration estimated in 1984 that manual 3-point safety belts reduce the fatality risk of front-seat occupants of passenger cars by 45 percent relative to the unrestra". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-253: NHTSA publishes 60% for front-seat occupants of light trucks (Kahane 2000, DOT HS 809 199). Our 95% interval [62.3, 68.4] does NOT cover it. Printed, not dropped.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/809199 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 253, dynamic-driver-ltv-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "16. Abstract The National Highway Traffic Safety Administration estimated in 1984 that manual 3-point safety belts reduce the fatality risk of front-seat occupants of passenger cars by 45 percent relative to the unrestra". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-258: NHTSA publishes 45% for front-seat occupants of passenger cars (Kahane 2000, DOT HS 809 199). Our 95% interval [43.1, 50.2] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/809199 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 258, dynamic-rf-car-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "16. Abstract The National Highway Traffic Safety Administration estimated in 1984 that manual 3-point safety belts reduce the fatality risk of front-seat occupants of passenger cars by 45 percent relative to the unrestra". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-263: NHTSA publishes 60% for front-seat occupants of light trucks (Kahane 2000, DOT HS 809 199). Our 95% interval [53.5, 60.2] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/809199 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 263, dynamic-rf-ltv-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "16. Abstract The National Highway Traffic Safety Administration estimated in 1984 that manual 3-point safety belts reduce the fatality risk of front-seat occupants of passenger cars by 45 percent relative to the unrestra". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-268: NHTSA publishes 54% for 3-point belts, outboard rear seat, passenger cars (NHTSA 2017, DOT HS 812 369 §1.6). Our 95% interval [49.8, 60.7] covers it.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 268, dynamic-rear-car-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "This situation may be changing. As of 2012, 27 States and the District of Columbia had laws requiring the use of seat belts by rear seat passengers of all ages – although in 10 of those States, they were “secondary” laws". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-273: NHTSA publishes 75% for 3-point belts, outboard rear seat, LTVs (NHTSA 2017, DOT HS 812 369 §1.6). Our 95% interval [57.6, 69.7] does NOT cover it. Printed, not dropped.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 273, dynamic-rear-ltv-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "This situation may be changing. As of 2012, 27 States and the District of Columbia had laws requiring the use of seat belts by rear seat passengers of all ages – although in 10 of those States, they were “secondary” laws". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-289: FHWA VM-1 publishes motorcycle travel as its own column. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 289, condition-note-0. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-290: FHWA VM-1, light-duty vehicles, short and long wheelbase. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 290, condition-note-1. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-291: FHWA VM-2, rural subtotal. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 291, condition-note-2. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-292: FHWA VM-2, urban subtotal. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 292, condition-note-3. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-293: FHWA VM-2, interstate columns. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 293, condition-note-4. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-294: FHWA VM-2, rural major collector. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 294, condition-note-5. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-295: FHWA VM-2, urban other freeways and expressways. Both halves of this fraction are published, so the answer is a measurement, not a choice.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 295, condition-note-6. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-300: The closest published figure is driver PREVALENCE at the roadside: the 2013-14 National Roadside Survey found 1.5% of weekend-nighttime drivers and 0.4% of weekday-daytime drivers at 0.08 or above. That is a share of drivers passing a sampled point, weighted by vehicle count. A driver on a 200-mile trip and one on a 2-mile trip count the same.
  https://rosap.ntl.bts.gov/view/dot/2084/dot_2084_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 300, condition-note-9. The source record's excerpt is a bibliography / sample-completion paragraph, which does not support these claims. Reading the report itself does: Table ES-1 gives "Weekday Daytime" 0.4 and "Weekend Nighttime" 1.5 at >=0.08. Its methods say the strategy "was used to weight the number of driver trips". This is sampled prevalence, not a national mileage census.
• CONFIRMED N2-302: NHTSA's observational survey puts belt use at 91.3% in 2025, defined as the share of front-seat occupants belted "during an average daylight moment". That is a share of people at a moment, in daylight. It is not a share of miles, and night is exactly when belt use is thought to fall.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813799 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 302, condition-note-11. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The national estimate of seat belt use by adult front-seat passengers of passenger vehicles in 2025 was 91.3 percent, not statistically different (at the .05 level) from 91.2 percent observed in 2024. This estimate repre". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-303: NHTSA's observational survey puts belt use at 91.3% in 2025, defined as the share of front-seat occupants belted "during an average daylight moment". That is a share of people at a moment, in daylight. It is not a share of miles, and night is exactly when belt use is thought to fall.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813799 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 303, belt-moment-quote. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The national estimate of seat belt use by adult front-seat passengers of passenger vehicles in 2025 was 91.3 percent, not statistically different (at the .05 level) from 91.2 percent observed in 2024. This estimate repre". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-307: 6 p.m. to 5:59 a.m., NHTSA's own clock
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813815 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 307, night-clock. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Weekday* 34,844 5,274 15% 33,900 5,194 15% Daytime 21,027 1,749 8% 20,607 1,750 8% Nighttime 13,719 3,486 25% 13,193 3,410 26% Weekend* 23,208 6,425 28% 21,647 6,148 28% Daytime 7,532 1,044 14% 7,102 970 14%". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-312: You grip the armrest at takeoff and let your guard down behind the wheel, but for the same point-to-point trip the steering wheel is the part that should scare you.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 312, summary-0-0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-315: A motorcycle rider carries under one percent of the traffic and more than fifteen percent of the deaths, over twenty times the average per mile, both halves published.
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 315, summary-0-4. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-321: The alcohol denominator was measured too, by a study that sampled the traffic instead of counting the miles.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 321, summary-0-10. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-322: For the same journey, the drive is far more dangerous than the flight, so the fear most people feel at takeoff is pointed at the wrong vehicle.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 322, summary-1-0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• WRONG N2-323: The page measures risk in micromorts, one micromort being a one-in-a-million chance of death, and recomputes every figure live from published death counts.
  https://en.wikipedia.org/wiki/Micromort ; public/strata/the-trip-that-flips-the-fear/index.html : The checker confirmed the micromort definition, but its source did not support the raw-count provenance clause. The calculator uses published rates plus a chosen floor. Corrected the plain frontmatter copy with X52/X55.
• CONFIRMED N2-325: There, the car must choose whom to risk; here, the choice is yours \u2014 fly or drive \u2014 and the per-mile death rate says the drive is the gamble.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 325, summary-2-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-339: Here NHTSA's 2023 road death count went from 40,901 to 41,025 when the Annual Report File was replaced by the Final File, at the same download address, because toxicology and death certificates arrive late.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/813791 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 339, summary-6-2. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The final fatality count in motor vehicle traffic crashes for 2023 was 41,025, updated from 40,901 in the 2023 ARF.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-342: For the same point-to-point trip, that is exactly backwards: per mile travelled, the drive accumulates far more **micromorts** \u2014 one-in-a-million chances of sudden death \u2014 than the flight.
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 342, summary-8-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-351: Road-safety epidemiology, facing the same hole, built estimators that do not need miles. **Double pair comparison** (Evans 1986, and NHTSA's standard belt method since the mid-1980s) puts two people in one crash and lets the second one's death rate stand in for how bad the crash was, so everything the two shared cancels and no exposure data is required at all.
  https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812369 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 351, summary-11-3. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Since the mid-1980s, fatality-reducing effectiveness estimates for occupant protection requiring activation (namely, buckling up), such as seat belts or child safety seats, have usually been based on double-pair comparis". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-353: And for alcohol, somebody went and measured the denominator by sampling the traffic itself, two control drivers per crash driver at the same place, same direction, same weekday and same hour, one week later.
  https://rosap.ntl.bts.gov/view/dot/1993/dot_1993_DS1.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 353, summary-11-5. Read DOT HS 812117 methods and Tables 7-8. Recruitment ran 20 months, all hours; controls matched location, direction, weekday and hour one week later. Table 7 gives crash N=3353, controls N=6868, BrAC>0.08 counts 95 and 26. The crude odds ratio is 7.6735. Adjusted risks at .05/.08/.10/.15/.20+ are 2.07/3.93/5.64/12.18/23.29. The page's "about 5.5" is approximate, not a verbatim table transcription.
• CONFIRMED N2-355: You white-knuckle takeoff and relax behind the wheel, but for the same trip the steering wheel is the part that should scare you.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 355, summary-13-0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-358: For the same trip, the drive piles up far more micromorts than the flight.
  https://usafacts.org/articles/is-flying-safer-than-driving/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 358, summary-14-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "The 2023 death rate for air passengers was 0.003 deaths per 100 million miles. The rate for car and truck passengers was 0.53.". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-362: Every click you hear is one micromort, a one-in-a-million chance of sudden death.
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 362, summary-18-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-370: Every click is one micromort, a one-in-a-million chance of sudden death, struck at whatever rate each mode's published death rate works out to under one exchange rate shown on screen.
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 370, summary-19-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-388: One click. One micromort: a one-in-a-million chance of sudden death.
  https://en.wikipedia.org/wiki/Micromort : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 388, film-cap-0. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "A micromort (from micro- and mortality ) is a unit of risk defined as a one-in-a-million chance of death .". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-389: Howard 1980 · Spiegelhalter, BJOG 2014
  https://obgyn.onlinelibrary.wiley.com/doi/10.1111/1471-0528.12663 ; https://plato.stanford.edu/archives/win2019/entries/rationality-normative-utility/ : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 389, film-citation-0. Curl received a challenge, but the publisher article was readable through web open: "DJ Spiegelhalter", "The power of the MicroMort", "Volume121, Issue6", "May 2014", "Pages 662-663". The Howard half of the film citation is independently supported by SEP.
• CONFIRMED N2-390: Ride with the car. Savage measured every US mode the same way, and the car costs 7.28 deaths for every billion passenger-miles.
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 390, film-cap-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Table 2 Passenger fatalities per billion passenger miles 2000e2009. Riding a motorcycle Driving or passenger in a car or light truck Passenger on a local ferryboat Passenger on commuter rail and Amtrak Passenger on urban". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-391: Savage 2013, Table 2 · US 2000 to 2009
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 391, film-citation-1. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Research in Transportation Economics 43 (2013) 9e22 Contents lists available at SciVerse ScienceDirect Research in Transportation Economics journal homepage: www.elsevier.com/locate/retrec Comparing the fatality risks in". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-395: Savage 2013, Table 2
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 395, film-citation-3. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Research in Transportation Economics 43 (2013) 9e22 Contents lists available at SciVerse ScienceDirect Research in Transportation Economics journal homepage: www.elsevier.com/locate/retrec Comparing the fatality risks in". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-399: Savage 2013, Table 2 · the rate is the pitch
  https://faculty.wcas.northwestern.edu/ipsavage/436.pdf : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 399, film-citation-5. Manually opened the source, including PDF text where the automated check was UNREADABLE. Matching source text: "Research in Transportation Economics 43 (2013) 9e22 Contents lists available at SciVerse ScienceDirect Research in Transportation Economics journal homepage: www.elsevier.com/locate/retrec Comparing the fatality risks in". Read in its surrounding section; source details are in the source review below.
• CONFIRMED N2-406: FARS 2024 · FHWA VM-1 and VM-2
  https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm1.cfm ; https://www.fhwa.dot.gov/policyinformation/statistics/2024/vm2.cfm ; git show 9b112ed892:research/the-trip-that-flips-the-fear/fars-counts.json : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 406, film-citation-9. The supplied excerpt covers only one side of this compound claim. Read FHWA VM-1/VM-2 jointly with the pinned FARS counts and original CSV. For 2024: motorcycle travel 22,241 of 3,294,031 million VMT and 6,225 of 39,254 deaths yields 0.675%, 15.858%, and 23.49x. Across all 14 road classes the extreme rate ratio is 4.495 in 2024 and 4.656 in 2023.
• CONFIRMED N2-413: Howard 1980 · Spiegelhalter, BJOG 2014
  https://plato.stanford.edu/archives/win2019/entries/rationality-normative-utility/ ; https://obgyn.onlinelibrary.wiley.com/doi/10.1111/1471-0528.12663 : Carried from the checker source reread in research/assay-line/tier-a-1001/the-trip-that-flips-the-fear/N2-codex.md. Entry 413, film-definition-attribution. SEP says "Howard (1980) introduces the concept of a micromort" and defines the one-in-a-million unit. The record's SEP excerpt does not mention Spiegelhalter; his own BJOG 2014 article, read via the web tool after curl 403, supplies that separate attribution.

What was done

• [fixed] X1: USAFacts credits BTS for the 2023 rates 0.53 and 0.003. Its air population includes on-demand air taxis; NSC has a separate scheduled-airline series. Attribution and population are now explicit.
• [fixed] X2: The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• [fixed] X3: The named speeds and Savage rates yield 0.25480 car and 0.03500 air micromorts/hour, a 7.28-fold difference. The prose now gives that ratio and requires leg durations for an airport-drive comparison.
• [fixed] X4: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X5: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X6: The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• [fixed] X7: DOT HS 813791 gives the old 40,901 and final 41,025. (41025-40901)/41025 = 0.003022547, about three thousandths or 0.30%. Corrected that decimal-scale description.
• [fixed] X8: DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• [fixed] X9: DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• [fixed] X10: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X11: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X12: double-pair.json gives driver-ltv 65.3% against 60%, but rear-ltv 64.0% against 75%. Corrected the claim that both misses run high; the root verifier checks both directions.
• [fixed] X13: DOT HS 812369 p.11 (PDF p.19) describes direct observation for some mortally injured occupants and over-reporting by survivors. The blanket claim that nobody saw belt use was narrowed.
• [fixed] X14: DOT HS 812369 includes the model-year 2014/2015 BODY_TYP exception to VIN decoding. The page now names that exception and treats selection differences as a hypothesis, not an established cause of the misses.
• [fixed] X15: The standalone research verifier checks arithmetic without reading the page. The root harness reads the conditioning and belt panels using artifacts and Playwright. The paragraph now states their separate coverage and requirements.
• [fixed] X18: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] X19: The original film log records 127 assertions, not 120. The caption now distinguishes the original 36/77 sign-off from the current authoring-source gate and states that the MP4 has not changed.
• [fixed] X20: USAFacts describes its 2024 0.19 transit-rail and 0.09 bus figures as passenger fatality rates. They are not totals inflated by trespassers. The explanatory claim was removed and the differing rail populations are named.
• [fixed] X21: USAFacts explicitly includes on-demand air taxis in its 2023 air series. Relabelled the 0.003 series wherever used; the separate Savage 2000–2009 commercial-aviation table retains its own scope.
• [fixed] X22: USAFacts explicitly includes on-demand air taxis in its 2023 air series. Relabelled the 0.003 series wherever used; the separate Savage 2000–2009 commercial-aviation table retains its own scope.
• [fixed] X23: 0.53/0.003 is driving divided by flying, 176.7. Only the inverted label was changed.
• [fixed] X24: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X25: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X26: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X27: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X28: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X29: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X30: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X31: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X32: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X33: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X34: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X35: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X36: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X37: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X38: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X39: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X40: DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• [fixed] X41: DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• [fixed] X42: DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• [fixed] X43: DOT HS 812369 specifies E = 100 - UEF * (100 - raw E). double-pair.mjs already implemented that residual-risk adjustment. Replaced the false displayed division, preserving the calculated estimates, and checked all four front-seat panels.
• [fixed] X44: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X45: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X46: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X47: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X48: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [fixed] X49: FHWA VM-1 cross-tabulates light-duty travel and highway category; 2024 rural/urban interstate cells are 197,623 and 506,376 million VMT. The page now uses that split; only its conversion to person-miles assumes national occupancy. Removed the universal absence and personal-benefit claims.
• [fixed] X50: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim.
• [dated] X51: Changed Article dateModified from 2026-07-26 to 2026-10-01. No Updated line is needed for metadata moved with the substantive corrections.
• [fixed] X52: The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• [fixed] X53: DOT HS 810948 Tables 2 and 3 use 2001 NHTS travel shares and 2006 FARS fatalities; approximately 25% of travel is at night. Table 3 reports 18.56 deaths of passenger-vehicle occupants aged 16+ per 100 million VMT for midnight to 3 a.m. The page now dates that older evidence, offers the 25% share as a sensitivity assumption, and denies neither its existence nor the lack of a matching 2023/2024 denominator here.
• [fixed] X54: DOT HS 813791 defines fatal motor-vehicle traffic crashes on public trafficways with death within 30 days. Raw FARS records contain unknown restraint and unmeasured BAC fields. Wording now states those limits.
• [fixed] X55: The inline calculator multiplies the published 0.53/0.003 rates by distance and adds a chosen 0.013-micromort flight floor. Raw FARS/FHWA counts feed the separate road panel. Corrected the prose, frontmatter, metadata and body copies.
• [fixed] X56: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] X57: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] X58: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] X59: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] X60: An independent Python CSV recount with INJ_SEV=4, PER_TYP in {1,2}, BODY_TYP 1..49 gives 24,025 occupants in 2023 and 22,708 in 2024. Known-restraint totals are 21,361 and 20,143; unrestrained totals are 10,531 and 9,751. Fixed the extractor, regenerated artifacts and data literals, and corrected the population-dependent page/film-source calculations. This exact scope is not asserted to equal NHTSA Table 4 totals of 24,017/22,713. The interstate rung now uses matching restricted deaths and light-duty VMT with national occupancy explicitly assumed, without a causal claim. The existing encoded film was not rendered or re-encoded. Its incorrect figures remain in the media and are explicitly listed as not done; corrected authoring sources and a precise visible note beside the player supply the permitted remedy.
• [fixed] N2-136: Corrected the overbroad surveyed-badge definition with X4; the roadside survey remains described as driver prevalence.
• [fixed] N2-323: Corrected the raw-count claim in plain frontmatter with the raw-inputs group; kept the micromort definition.