Why a One-Truck Owner-Operator Can Be Safer Than a 50-Truck Fleet (and How to Tell at a Glance)

The most useful safety conversation I've had with another broker in the last year started with him saying "we just don't book single-truck operators anymore. Too risky." When I asked him to show me his criteria, he showed me a spreadsheet that effectively filtered out any carrier with fewer than five power units. He was solving for what he thought was risk. He was actually solving for "I don't want to look at the file carefully," and screening out a whole population of carriers who, in many cases, are operating cleaner than the mid-size fleets he was happy to book.

This post is about why the instinct is backwards, and what the metrics that actually predict risk look like once you adjust for fleet size.

The instinct vs the data

The instinct: bigger fleet = more drivers = more management = more safety culture = safer.

The data, in pieces:

FMCSA's own analysis showed that motor carriers within 24 months of graduating the new-entrant program had roughly twice the crash rate of established carriers — and most new-entrant graduates are small fleets. So far so good for the instinct.

But the SAME data, examined per-power-unit instead of total, shows that some single-truck operators with five+ years of authority were operating at crash rates **below** the established-carrier average. The new-entrant problem isn't a small-carrier problem. It's a new-carrier problem. Conflating them is what produces the lazy "bigger = safer" heuristic.

When you actually rank carriers by per-power-unit crash rate or per-power-unit OOS-inspection rate, the distribution does NOT cleanly correlate with fleet size. A well-run 2-truck operation, where the owner is also the driver and runs the truck like they own it (because they do), can outperform a 40-truck fleet where the fifth-best driver is dragging the average down.

The CSA BASIC system actually addresses this with its **peer-group methodology**. BASIC percentiles aren't computed against the whole industry — they're computed against carriers of similar size and operation. A 2-truck carrier with a 75% Unsafe Driving percentile is at 75% **among other 2-truck carriers**, not the whole industry. Similarly for the 50-truck fleet. The peer-group structure is FMCSA's way of saying "fleet size matters, but the comparison that matters is against your peers."

If you ignore peer-group context and just look at raw count of inspections, raw count of crashes, raw count of out-of-service violations — you'll systematically misjudge small carriers.

The metrics that actually correlate with operational risk

Three measurements I use, in order of importance:

1. Driver out-of-service rate (DOOS). This is the percentage of driver inspections that resulted in the driver being put out of service. The national average is around 5.5%. A carrier of any size that's running 3% DOOS is operating cleaner than the national average. A carrier of any size at 12% DOOS has a real driver-qualification or hours-of-service problem. The metric is per-inspection, so it adjusts for the fleet-size question automatically.

2. Vehicle out-of-service rate (VOOS). Same idea, but for the vehicle. National average around 20.7%. A carrier at 14% is maintaining the equipment better than peers. A carrier at 35% is putting unsafe trucks on the road. This is the single best leading indicator of a mechanical-failure crash that I know of.

3. Crashes per power unit (CPU). Take the carrier's 24-month crash count and divide by their power-unit count. A 2-truck carrier with one preventable crash has a CPU of 0.5 — that's a high CPU. A 100-truck carrier with five preventable crashes has a CPU of 0.05 — that's a low CPU. The 100-truck fleet has more total crashes, but their per-truck crash exposure is an order of magnitude lower.

CPU is rough — it doesn't adjust for mileage, equipment type, or geography — but it's a useful sanity check against raw crash count.

When all three of those run in a small carrier's favor, the carrier is genuinely operating better than peer averages and possibly better than the mid-size fleets you'd default to. When all three run against them, the carrier has a real problem regardless of size.

A concrete scenario

You're vetting two carriers for the same lane.

Carrier X: MC-1402189 / DOT-3712840. Two power units, three drivers. Authority granted seven years ago. 14 inspections in the last 24 months. Two driver OOS events (DOOS 14.3%). One vehicle OOS (VOOS 7.1%). One crash, non-preventable (rear-ended at a red light). Insurance L&I clean.

Carrier Y: MC-1289304 / DOT-3019847. 35 power units, 41 drivers. Authority 12 years. 410 inspections in the last 24 months. 31 driver OOS events (DOOS 7.6%). 91 vehicle OOS (VOOS 22.2%). Four crashes, three preventable. Insurance L&I clean.

The instinctive read: Y is bigger and more "established," so the safer bet. Y has more drivers under more supervision, has been around longer, runs bigger volume.

The data-driven read: Y's DOOS is above the 5.5% national average. X's DOOS is significantly above national average too — but on a much smaller sample (2 OOS events out of 14 inspections; small-sample noise). Y's VOOS at 22.2% is worse than national; X's at 7.1% is materially BETTER. Y has three preventable crashes in 24 months on 35 trucks (CPU of 0.086 preventable). X has zero preventable crashes (CPU of 0).

You can't conclude X is safer with high confidence — the sample size is tiny. But you can confidently conclude that "Y is bigger so Y is safer" is not supported. If your file says you picked Y because they had more trucks, you've documented exactly the kind of unsophisticated diligence a plaintiff's lawyer wants to find in your records.

The right framing for small carriers

Don't auto-decline single-truck or two-truck operators. **Vet them harder, and document the additional vetting.** What "harder" means in practice:

• Verify owner identity and prior authority history (anti-chameleon check).
• Look at how long the owner has held a CDL — many small carriers are run by drivers with 15+ years of experience who just decided to go independent.
• Look at the maintenance evidence — annual inspection records, recent roadside inspections with clean results.
• Look at the named insured on the BMC-91 — does it match the operating entity exactly.
• Look at MCS-150 — is the carrier's reported operation consistent with what they're quoting.
• Phone call with the owner. You learn more about a carrier's safety culture in a 4-minute phone call than from any document. Is the owner the driver? How do they handle pre-trip on the rare days they hand off to another driver?

The brokers who say "I don't book single-truck" are leaving capacity on the table — there are tens of thousands of competent single-truck operators in the US — and they're masking a vetting process they haven't actually built.

The regulation, in plain English

CSA's peer-group methodology is documented in the FMCSA SMS Methodology document. The intervention thresholds vary by fleet size as well as carrier type. For very small fleets (Safety Event Group 1, generally <5 power units), some BASIC categories use the "Insufficient Data" treatment because the sample is too small to compute a reliable percentile. This means small carriers often have FEWER public percentile alerts than larger ones — not because they're safer, but because there isn't enough data to score them on.

For the broker file, this is a place where "Insufficient Data" needs to be explicitly handled. A 2-truck carrier with no BASIC percentiles isn't a safe carrier OR an unsafe carrier — they're an unknown. The diligence weight shifts to the things that ARE measurable: per-inspection OOS rates, preventable-crash count, the named-insured on the insurance filing, the authority age, and the substance of the owner phone call.

How I document this

For any carrier with fewer than 5 power units, the file adds three lines beyond the standard packet:

1. **Owner identity and prior authority history**, captured from a phone call with the carrier and verified against the BMC-91 named insured.

2. **Per-inspection OOS rates** computed and noted (DOOS%, VOOS%), with the comparison to national averages (driver ~5.5%, vehicle ~20.7%) written out.

3. **A rationale paragraph** that explicitly addresses "we vetted this small carrier because the per-truck metrics outperformed the peer-group benchmarks" — or "we vetted and passed because the per-truck metrics did not."

That paragraph is the difference between "broker decided to book a 2-truck carrier without thinking about it" and "broker reviewed the size-adjusted operational data and made a documented call." The first is the negligent-selection plaintiff's argument. The second is the broker's defense.

— Mason Lavallet

Founder, DOTScreener.com

---

Sources

• [FMCSA SMS Methodology (PDF)](https://csa.fmcsa.dot.gov/Documents/SMSMethodology.pdf) — peer-group methodology + safety event groups
• [FMCSA — Compliance, Safety, Accountability program overview](https://csa.fmcsa.dot.gov/about/Measure)
• [FreightWaves — FMCSA data shows rise in crash rates among new-entrant carriers](https://www.freightwaves.com/news/fmcsa-data-shows-rise-in-crash-rates-aomng-new-entrant-carriers)
• [American Transportation Research Institute (ATRI) — small motor carrier research](https://truckingresearch.org/)
• [49 CFR Part 385 Subpart D — New Entrant Safety Assurance Program](https://www.ecfr.gov/current/title-49/subtitle-B/chapter-III/subchapter-B/part-385/subpart-D)