The Crash Indicator BASIC Counts Crashes, Not Fault. Here's Why It Still Matters.

The BASIC Nobody Reads Right

A carrier came through our pipeline a few years back — DOT-3421578, MC-1193472, 4-year-old authority, 12 trucks out of Tennessee. Clean Unsafe Driving BASIC. Clean HOS. Clean Vehicle Maintenance. But the Crash Indicator was sitting at the 79th percentile.

I flagged it. My ops person pushed back: "Their crashes weren't their fault. One was a rear-end where someone hit them, one was a blown tire from road debris." He wasn't wrong about the specifics. But I asked him: how did he know that? He said the carrier told him.

That's not reading the Crash Indicator BASIC. That's reading the carrier's explanation of it.

What the Crash Indicator BASIC Actually Measures

The Crash Indicator is one of seven Safety Measurement System categories FMCSA publishes in SMS. Unlike the other six — which measure inspection violations — this one measures crash involvement. Specifically, it measures crashes per vehicle mile traveled over a rolling 24-month window.

The crashes feeding that BASIC come from the Motor Carrier Management Information System, or MCMIS. Police-reported crashes involving commercial vehicles get entered by state DOTs. FMCSA aggregates them. The denominator — the VMT — comes from the carrier's most recent MCS-150 filing.

Two things matter immediately.

First: MCMIS counts crashes you were involved in, not crashes you caused. If a distracted driver in a sedan rear-ends your carrier's loaded flatbed at a red light, that crash goes into MCMIS. The carrier's Crash Indicator goes up. There is no fault field in the Crash Indicator BASIC. The SMS doesn't distinguish.

Second: the VMT denominator is self-reported. The carrier fills out the MCS-150 (required at minimum every two years under 49 CFR § 390.19), estimates their annual miles, and FMCSA uses that figure to calculate crashes-per-VMT. A carrier that underestimates their VMT — and this happens often, either deliberately or because tracking actual miles across a fleet is tedious — ends up with a higher Crash Indicator than their actual exposure justifies. A carrier that inflates their VMT looks cleaner than they are.

This denominator issue is real and it matters, but it's not a reason to dismiss an elevated Crash Indicator. It's a reason to cross-check it.

Why a High Percentile Still Means Something

Here's where I part ways with people who use the "it wasn't their fault" argument to dismiss an elevated Crash Indicator entirely.

Even granting that specific crashes weren't the carrier's fault, a high crash-involvement rate still tells you something. High crash involvement — regardless of fault — suggests the carrier operates in conditions where crashes keep finding them. That can reflect poor route selection, running high-density corridors during peak hours, or drivers who don't practice the kind of defensive awareness that keeps other people's mistakes from becoming your crashes.

It can also mean their drivers are excellent defensive operators who keep getting unlucky. That's possible. But you don't know which one it is without digging.

The Crash Indicator is what tells you to dig.

Compare this to the Unsafe Driving BASIC: that BASIC measures violations — speeding, improper lane change, following too close — which are direct observations of driver behavior. It tells you what inspectors saw. The Crash Indicator tells you what happened. Both matter, but differently. A carrier can have a clean Unsafe Driving BASIC and a high Crash Indicator, which means inspectors haven't caught bad behavior but crashes are still occurring. That combination is worth a harder look, not a pass.

The Problem at Tender Time After Montgomery

Montgomery v. Caribe Transport II changed how this analysis sits in a broker's file. The Supreme Court held unanimously in May 2026 that state-law negligent selection claims against freight brokers aren't preempted by federal law. Brokers can be sued. Files get subpoenaed. People get deposed.

If you tender a $1.2M electronics load to a carrier sitting at the 82nd percentile Crash Indicator, a deposition question is coming: "Were you aware this carrier had crash involvement in the top 20% of comparable carriers in the FMCSA Safety Measurement System?" You'd better have a documented reason for that tender.

"The carrier explained the crashes" is not documentation. "We reviewed the SMS Crash Indicator at the 82nd percentile, called the carrier, confirmed two of three crashes in the 24-month window involved third-party at-fault drivers per their accident reports, and the third was a road-hazard incident with a corresponding insurance file — a pattern inconsistent with driver-fault crash behavior" is documentation.

One is a phone call you can't prove happened. The other is a record that can go in front of a jury and hold up.

Reading the Crash Register, Not Just the Percentile

The Crash Indicator BASIC gives you the percentile. The accident register gives you the circumstances.

Under 49 CFR § 390.15, motor carriers are required to keep a record of each accident involving their commercial motor vehicles for at minimum three years. The register must include the date, location, vehicle identification, driver name, number of injuries, number of fatalities, and whether any hazmat was released. It must be made available to FMCSA on request.

The plain-English version at load-tender time: the carrier is legally required to have this record. If you ask for it and they say they don't have one, they're either non-compliant or they're hoping you won't call them on it. Either answer is informative.

What you can see from SAFER without calling anyone: the SMS Safety Events breakdown shows the number of crashes in the 24-month window and categorizes them as fatal, injury, or tow-away. A carrier with three tow-away crashes is a different conversation than a carrier with three minor fender-benders where the other driver filed a police report. Fatality and injury crashes carry orders of magnitude more liability exposure; tow-aways at least tell you a vehicle was damaged enough to need a flatbed.

The SAFER company snapshot also shows whether a carrier has had compliance reviews. A carrier who's been formally reviewed by FMCSA — especially more than once — and still carries an elevated Crash Indicator is a carrier FMCSA has looked at and either cleared or flagged. Check the review outcome on the SAFER page. A carrier with a "Satisfactory" rating following a compliance review is in a different bucket than one who's never been reviewed at all.

The VMT Problem in Practice

I want to come back to the denominator because it cuts both ways.

Small carriers and newer authorities with limited VMT have erratic Crash Indicator percentiles. If you're looking at a six-truck operation that's been running 18 months, one crash in MCMIS can swing the percentile significantly. The "Insufficient Data" flag that FMCSA applies in SMS is there because small-sample percentiles are unreliable signal. When you see "Insufficient Data" on the Crash Indicator, don't read it as clean. Read it as unknowable from this metric.

For mid-size fleets — 15 trucks and up with a few years of operation — the Crash Indicator percentile stabilizes into something more meaningful. The VMT estimate matters more, and you can at least ask the carrier: "When did you last file your MCS-150, and was the mileage you reported for last year accurate?" A carrier who knows their MCS-150 filing date and can give you a rough mileage figure has a functioning compliance mindset. A carrier who doesn't know what an MCS-150 is needs a different conversation.

What I Actually Do With a High Crash Indicator

At the 65th percentile or above, I look at the SMS Safety Events breakdown to count the underlying crashes and see whether they're tow-away, injury, or fatal. I check the SAFER snapshot for compliance review history.

Above the 72nd percentile, I call the carrier and ask: "Tell me about the crashes on your FMCSA record in the last two years." I'm not asking them to exonerate themselves. I'm listening for whether they can describe the events accurately, whether they've made any operational changes in response, and whether they try to wave it off as someone else's problem without specifics.

At or above the 80th percentile, I want documentation before I tender. A brief written summary of each incident, any third-party accident reports they can share, and whatever their insurance carrier has on file. None of this is FMCSA-required, but all of it is reasonable to request when a metric is elevated and a claim exposure is real.

What I won't do is take a carrier at the 85th percentile Crash Indicator and tender a load with $800K in cargo value because they seemed trustworthy on the phone and I ran out of time before dispatch. That's the file that hurts in discovery.

How I Document This

In the carrier record, I note the Crash Indicator percentile at the time I ran the check, the date I ran it, and the underlying crash count from SMS. If the percentile is elevated, I add:

• Number of crashes in the 24-month window and their classification (fatal/injury/tow-away)
• Date of the carrier's most recent MCS-150 and whether VMT appears current
• Any compliance review history and outcome noted on SAFER
• What the carrier told me, verbatim or summarized, about the crashes and the date I spoke with them
• My decision: tendered because [specific reasoning], or declined because [specific reasoning]

If the Crash Indicator is under 60th percentile, I document that it was within acceptable range at time of tender and move on. I don't need a paragraph about a clean BASIC. But elevated BASICs require documented evaluation — not just a mental note that the carrier explained it away.

The Crash Indicator BASIC is the one carriers push back on most, because "it wasn't our fault" sounds reasonable. Sometimes it even is. But you can't enter that into your carrier file as an assertion. You have to show the work.

---

— Mason Lavallet

Founder, DOTScreener.com