The Carrier Had Five Crashes. None Were Their Fault. You're Still Getting Sued.

The defense that sounds reasonable until a lawyer starts asking questions

A broker I know had a carrier with five crashes over three years in their SAFER snapshot. He tendered a load. The load was involved in a serious accident. In discovery, plaintiff's counsel pulled the carrier's crash history and put it in front of the broker at deposition.

"Did you see that this carrier had five crashes in the three years before you hired them?"

"Yes, but none of them were their fault. FMCSA doesn't factor fault into the Crash Indicator score."

The plaintiff's attorney let that answer sit for a second. Then she asked: "And where in your file did you document that analysis?"

There was nothing in the file. He'd looked at the number, seen the score, told himself it didn't count, and moved on. No notes. No documentation of the reasoning. No indication that he'd done anything more than look at a number and dismiss it.

That gap — between what he thought and what he could prove — is where negligent selection cases live.

What FMCSA actually says about fault

The Crash Indicator BASIC includes crashes reported to FMCSA regardless of which party was at fault. A carrier can have a truck rear-ended at a stop sign four times and every one of those crashes goes into their Crash Indicator record. FMCSA is explicit about this. They use crash data as a proxy for exposure and safety culture, not as a finding of fault.

For purposes of the BASIC scoring methodology, fault is irrelevant. FMCSA has said so publicly.

That's true. And brokers who know this — and it's not a secret, it's in the SMS documentation — often use it as a reason to look past a carrier's crash record. If the Crash Indicator score is elevated but the crashes weren't the carrier's fault, does it really tell you anything useful?

Here's where the reasoning breaks down.

Negligent selection doesn't ask who caused the crash

After Montgomery v. Caribe Transport II (U.S. Supreme Court, May 2026), freight brokers can be held liable in state court for negligently selecting a motor carrier. The standard isn't "did the broker cause the accident?" The standard is whether a reasonably prudent broker, given the information available, made a reasonable selection decision.

What goes into that standard? Every data point the broker had access to. Including the carrier's crash history.

A carrier with five crashes over 36 months is a carrier with five documented crash events on their federal record. The plaintiff's attorney doesn't need to prove any of those crashes were the carrier's fault to use them against you. What they need to show is that a pattern existed in the data available to any broker doing reasonable due diligence — and that you either didn't look, looked and didn't document, or looked, dismissed it, and kept no record of why.

The "it wasn't their fault" reasoning is potentially legitimate. But it has to be documented at the time of selection, not reconstructed from memory at a deposition.

The composite picture that actually matters

A carrier's crash history doesn't exist in isolation. It's one data point among many, and it has to be read against the rest of what you know about the carrier.

Take a carrier: MC-1247893, DOT-3419872. Three years of authority. Five crashes in the last 36 months — all reported as rear-end collisions where the carrier was struck. Their Crash Indicator BASIC is at the 74th percentile for property carriers. Not over the 65-percent FMCSA intervention threshold. But their Unsafe Driving BASIC is at the 71st percentile. Their Vehicle Maintenance BASIC is at the 58th percentile. Their HOS Compliance BASIC is at the 62nd percentile.

None of those individual scores cross the intervention threshold. But three BASICs elevated in the 58th–74th range, combined with five documented crash events and an authority age of three years, paints a picture of a carrier operating near the edge on multiple safety dimensions simultaneously.

That's what a plaintiff's expert witness uses. Not any single data point. The composite.

And here's the part brokers miss: even if the crashes were genuinely not the carrier's fault, elevated BASICs in Unsafe Driving and HOS alongside a pattern of crash involvement creates a reasonable question about whether something systematic was happening at that carrier. Maybe they drove fatigued. Maybe their drivers were speeding when someone cut them off. Maybe their pre-trip inspections missed equipment issues that contributed to crash outcomes. FMCSA's fault determination doesn't settle any of those questions.

What actually goes in the crash record

The SAFER snapshot shows crash counts — fatal, injury, and property damage — over the last 24 months in the state inspections section, and the Crash Indicator BASIC roll-up on the SMS side. That's what most brokers look at.

What it doesn't show you directly: the geographic distribution of those crashes, the severity breakdown by specific event, whether any crash resulted in a driver cited for a moving violation, or whether any carrier was placed out of service at the scene.

Under 49 CFR § 390.15, motor carriers are required to maintain an accident register with specific details for each accident — location, date, driver, vehicle, injuries and fatalities, whether a citation was issued. You can't see that register without a request to the carrier. But the SAFER data gives you enough to know whether a more detailed inquiry is warranted.

Five crashes in 36 months with no other elevated signals might be reasonable in context. Five crashes with three elevated BASICs warrants at minimum a documented decision and ideally a call to the carrier to understand their safety response.

The citation that matters for your file

49 CFR § 387.7 requires motor carriers to maintain financial responsibility on file with FMCSA. That's the insurance piece. But for crash history and the broader safety standard of care, the operative framework post-Montgomery is state negligence law — and what a state court will look at is whether your selection process was reasonable given what a broker with access to public FMCSA data should have seen.

The standard isn't "did you check the Crash Indicator BASIC." The standard is: what did you do with what you saw?

Separately, 49 CFR § 392.2 requires drivers to comply with state traffic laws, and Part 395 governs hours of service. Neither has a direct bearing on your carrier selection standard of care — but a carrier with repeated HOS violations in their BASIC record and repeated crash involvement is a carrier where the combination of data points should have prompted questions. A jury doesn't need a legal citation to understand that elevated fatigue-risk signals plus a crash pattern adds up to something a careful broker would have noticed.

The argument plaintiffs' attorneys make post-Montgomery isn't complicated: public data was available, you had access to it, you approved the carrier anyway, and something bad happened. The fault determination on any individual crash is a sideshow. The question is whether the aggregate picture in the public data was one a reasonable broker should have acted on — and if you can't show what you did with it, the answer defaults against you.

Why the "not at fault" argument gets harder over time

Here's the thing about repeated not-at-fault crashes. By the third one, a reasonable observer might ask why a carrier keeps being in situations where other vehicles crash into them. Are they parking in dangerous locations? Stopping suddenly? Operating in high-density corridors with above-average incident rates? Is there a pattern in when and where the crashes occur?

None of that proves fault. But a carrier that's had five crashes in three years — regardless of attribution — has had five opportunities for something to go wrong on one of your loads. That's a real data point about exposure, not just about safety culture.

And if you approved them without noting it, you approved them without any documented evidence that you thought about it at all.

What I document when the crash record is elevated

When DOTScreener pulls a carrier and their crash history is worth examining, here's what goes into my file:

The raw numbers from the screening record — crash count, severity breakdown, BASIC percentile, all timestamped. Then a brief written note on what I saw and how I processed it. If I'm relying on the "not at fault" reasoning, I say that explicitly — "Crash Indicator elevated, five events over 36 months, all reported as not-at-fault rear-end collisions; other BASICs within normal range; proceeding with approval." If I made a call to the carrier about their safety program in response to the crash history, I log the call — who I spoke to, what they said, when it happened.

The difference between a file that survives deposition and one that doesn't usually isn't the decision you made. It's whether you can show you made it on purpose, with information in front of you, and that you thought about what the information meant.

A crash record you looked at and dismissed without documentation isn't a clean file. It's an open question about whether you looked at it at all.

DOTScreener keeps the timestamp on every carrier screen so that question has a clear answer: you ran the carrier at this date and time, the crash data showed this, and the file reflects your decision. That's the layer of defensibility the "I check all my carriers" claim actually needs.

— Mason Lavallet

Founder, DOTScreener.com