Methodology

How fleetfax reads the federal safety record

Every fleetfax report is built from the public federal record. This page explains, in plain terms, where the numbers come from, what the safety percentiles mean, when one is not computed, and how to correct a number you think is wrong.

Every figure on a fleetfax report comes from public federal records, mostly FMCSA’s. fleetfax never tells you whether to book a carrier. It shows every fact and flag the record holds and leaves the judgment with you. There is no single safety score. One number cannot cover insurance, safety, authority, and crash history.

Where the data comes from

fleetfax reads the public federal record that FMCSA publishes about every registered carrier: registration and company details, operating authority, insurance filings, roadside inspections, crashes, and safety ratings. Every report states the date its safety data runs through. FMCSA publishes its safety measurement files monthly, so the safety percentiles can trail the registration and inspection record by several weeks.

Freshness is ultimately bounded by the source. FMCSA publishes different datasets on different cadences, and no tool built on the public record can be fresher than the record itself. Where a figure is tied to an older filing, the report says so.

The 24 federal datasets

fleetfax uses these 24 federal datasets. A note beside a dataset says when it stopped updating or how it is loaded. FMCSA datasets are published on data.transportation.gov, with the dataset id after each name.

FMCSA registration and authority

• Company Census, FMCSA az4n-8mr2
• Carrier registration (Motus), FMCSA inys-ebih
• Authority history (Motus), FMCSA yu5v-wbh6
• Authority history (legacy), FMCSA 9mw4-x3tu (no updates since May 14, 2026)
• Revocations and suspensions (Motus), FMCSA wb4f-neki
• Revocations (legacy), FMCSA sa6p-acbp (no updates since May 14, 2026)
• Out-of-service orders, FMCSA p2mt-9ige
• Licensing and insurance carrier file (legacy), FMCSA 6eyk-hxee (no updates since May 14, 2026)

Insurance

• Insurance on file (Motus), FMCSA c5y8-a4uz
• Insurance history (Motus), FMCSA 3uet-3z4i
• Active and pending insurance (legacy), FMCSA qh9u-swkp (no updates since May 14, 2026)
• Insurance history (legacy), FMCSA 6sqe-dvqs (no updates since May 14, 2026)

Inspections, violations and crashes

• Vehicle inspections, FMCSA fx4q-ay7w
• Inspections and violations, FMCSA 876r-jsdb
• Inspections per unit, FMCSA wt8s-2hbx
• Crashes, FMCSA aayw-vxb3

Safety measurement (SMS)

• SMS input: violations, FMCSA 8mt8-2mdr
• SMS input: inspections, FMCSA rbkj-cgst
• SMS results: interstate carriers and intrastate hazmat carriers, FMCSA 4y6x-dmck
• SMS results: passenger carriers, FMCSA m3ry-qcip
• SMS results: intrastate passenger carriers, FMCSA h3zn-uid9

Other federal sources

• VIN decoding (vPIC), NHTSA
• Enforcement cases, FMCSA ai.fmcsa.dot.gov
• Gazetteers and population estimates, U.S. Census Bureau (fixed reference data)

Rejected insurance filings (FMCSA 96tg-4mhf) and hazmat incident reports (PHMSA) were removed from this list on Oct 14, 2026: fleetfax no longer reads them.

The 24-month window

The safety signals in a report are computed over a rolling 24-month window, the most recent two years of the record. This is the same window FMCSA’s own Safety Measurement System (SMS) uses for its safety calculations, so a fleetfax number looks at the same recent slice of a carrier’s history that the federal system looks at.

Comparing carriers within peer groups

A raw count is hard to read: a large fleet racks up more inspections and more crashes simply by running more trucks. So the safety picture is expressed as a percentile, meaning where a carrier sits relative to other carriers rather than an absolute count.

FMCSA’s SMS sorts carriers into safety event groups so they are compared to peers with similar road activity, rather than the entire industry. fleetfax applies this same grouping to inspection BASICs and unsafe driving. Crash rate is the exception: it is measured per million miles. Because normalizing by mileage already removes the fleet-size bias, a carrier is ranked in one national pool rather than a size group: every carrier with at least two recordable crashes and recent, usable mileage.

What “fleetfax peer estimate” means

You will see fleetfax percentiles labeled as a fleetfax peer estimate. Under the FAST Act of 2015, FMCSA removed property carriers’ official BASIC percentiles from public view (though carriers can still see their own, and passenger-carrier percentiles remain public). fleetfax calculates its own peer estimate using the exact same public inspection and crash records FMCSA relies on. It is fleetfax’s best model of a number the agency keeps private. It is never an official CSA, SMS, or FMCSA score. Peer estimates are rebuilt from each monthly release of FMCSA’s safety measurement files.

When no percentile is estimated

fleetfax only calculates a peer estimate when a carrier has sufficient underlying public records to make the comparison meaningful. This requires enough recent inspections or crashes alongside up-to-date filing data, mirroring FMCSA’s own policy of establishing minimum activity thresholds before scoring carriers against peers. When a carrier’s record is too thin or outdated to support a reliable estimate, fleetfax displays the raw counts and underlying facts with no percentile attached.

The crash rate

fleetfax displays crashes as a rate per million miles rather than a raw count, allowing large and small fleets to be evaluated on the same scale. The rate divides recordable crashes in the 24-month window by two years of the carrier’s reported annual mileage. FMCSA counts a crash as recordable only if it involves a fatality, an injury requiring medical treatment away from the scene, or a vehicle towed from the scene.

One caveat: because mileage is self-reported on the carrier’s MCS-150 filing, the rate relies entirely on the accuracy of that submission. The crash count covers the last 24 months, but the mileage is the carrier’s last reported figure, often one to two years old. The report’s crash tile labels the rate “last 24 mo”, and its fleet section shows that mileage with the MCS-150 year it was reported for. For baseline context, reports show the national average FMCSA published in its Safe Driver Apprenticeship Pilot eligibility criteria: roughly 0.96 crashes per million miles. That is the industry average, not a pass-fail line; the report flags a rate at FMCSA’s 1.5 compliance-review threshold, or when a carrier’s rate stands well above its peers.

Fatal crashes are read the same way, by rate rather than by count. A fatal crash is on the record either way, and it carries its own flag, but a raw count says nothing about how much driving it came from. fleetfax divides fatal crashes in the 24-month window by the same mileage the crash rate uses and compares the result to the national figure for large trucks: roughly 1.63 fatal-crash involvements per 100 million miles driven (NHTSA, Traffic Safety Facts 2023, Large Trucks; FMCSA’s Large Truck and Bus Crash Facts publishes the same measure off a slightly different mileage source). It counts trucks involved in fatal crashes, across every truck over 10,000 pounds, so it is context, not a peer rate. The national rate would predict even half a fatal crash only at about 30 million miles over the window. Below that, fleetfax makes no comparison in either direction and says so, because at that mileage one crash is chance rather than a rate. The comparison sets how much weight the fatal crash carries. It is a caution when the carrier’s own fatal rate sits above the national figure, or when there are too few miles to compare, and a note when the rate sits at or below the national figure on enough miles to say so. Either way the fatal crash stays on the report.

A report is evidence, not a judgment

fleetfax surfaces facts and flags; it does not make the decision for you. The percentiles, the crash rate, and the risk flags are evidence you weigh. The judgment, and the responsibility for it, stays with you.

Report a data error

Because fleetfax mirrors the public federal record, you can always verify any figure against the source. If fleetfax differs from the live federal record, use the Report a data error link on the carrier report or email [email protected].

If the underlying federal record itself is wrong, submit a Request for Data Review through FMCSA’s DataQs portal. Once FMCSA corrects the record at the source, the fix reaches fleetfax when that file next updates. The site status page shows how fresh each carrier dataset is.