What Is a Good CSA Score? The Real Thresholds, BASIC by BASIC (2026)

Published August 21, 2026 | TruckerNavi Inc | 8 min read

The short answer: there is no good CSA score, because there is no a CSA score. FMCSA scores seven categories separately, each on a 0 to 100 percentile where higher is worse, and sets a different intervention threshold for each. For a general freight carrier the line is 65% in Unsafe Driving, Crash Indicator and HOS Compliance, and 80% in the other four. Passenger carriers are held to 50% and 65%, placardable hazmat carriers to 60% and 75%, with HM Compliance the one exception: 80% for every carrier type. Anything at or above your threshold flags you. Everything below it is, operationally, a good score.

The second half matters more for a small fleet: a percentile is not a grade, it is a ranking against carriers of roughly your own size. That single design choice is why one bad inspection moves a 3-truck carrier twenty points and a 300-truck carrier not at all. Numbers below come from FMCSA's SMS Methodology (June 2026) and the agency's SMS site. For the whole system end to end, start with our CSA Scores Explained guide.

What are the CSA intervention thresholds for each BASIC?

This is the table people are actually looking for. FMCSA groups the seven BASICs into three tiers and applies three sets of thresholds depending on what kind of carrier you are. A lower threshold means stricter, because higher percentiles are worse.

• The strict tier is not random. FMCSA states that Unsafe Driving, Crash Indicator and HOS Compliance have the strongest associations to crash risk, so they get the tightest thresholds. If you can only fix one thing, fix there. What a single logbook violation costs is broken down in HOS violations and CSA points.
• It is "at or above", not "above". A general carrier sitting exactly at 65.0 in HOS Compliance is flagged, not borderline.
• Your tier is decided by data, not opinion. Hazmat thresholds apply if you had at least 2 placardable vehicle inspections in 24 months, at least 1 in 12 months, and at least 5% of your vehicle inspections were placardable, or if you hold a Hazardous Materials Safety Permit. Passenger thresholds need active passenger authority in L&I plus at least 2% of your power units being 9+ passenger capacity vehicles; or a for-hire operation type on the MCS-150 with that same 2% test (or no power units registered at all with passengers listed as cargo); or a private-passenger operation type with at least 2% of power units at 16+ capacity. Everyone else is general.

What does the percentile actually mean? It is a ranking, not a grade

FMCSA calculates a BASIC measure for you, lines up every comparable carrier's measure in ascending order, and converts positions into percentiles from 0 to 100. Higher means worse relative to that lineup. The measure is a fraction, and the denominator is not the same in every BASIC:

Two BASICs are normalized by fleet size, five by how often you get inspected. A carrier that runs clean but is constantly pulled into scales builds a large denominator in those five, which cushions every violation. A rarely inspected carrier has no cushion. Average power units are themselves an average of three snapshots: your current count, your count six months ago and your count eighteen months ago, divided by three.

Severity weights run from 1 (less severe) to 10 (most severe), and every violation carries a time weight on top: 3 in the first 6 months, 2 from 6 to 12 months, 1 from 12 to 24 months, then out of the calculation entirely (see when CSA violations fall off).

And the fact that catches almost every new owner: violations found on an inspection count whether or not a citation or ticket is issued. The inspection report is the input, not the courtroom outcome.

Who exactly are you compared to? Safety event groups

You are not ranked against all carriers in the system. FMCSA sorts carriers into safety event groups and ranks you only inside your group, but what gets counted differs by BASIC: inspections that carried a violation in Unsafe Driving and Controlled Substances/Alcohol, crashes in Crash Indicator, and all relevant inspections including clean ones in the other four. Unsafe Driving and Crash Indicator also split carriers into a Combination segment (combination trucks or motor coaches at 70% or more of power units) and a Straight segment (straight trucks and other vehicles over 30%).

Below a minimum you get no percentile at all, which is not a punishment: FMCSA simply does not consider the sample large enough. The minimums split two ways. Unsafe Driving needs 3 inspections that carried an Unsafe Driving violation, Controlled Substances/Alcohol needs 1, and Crash Indicator needs 2 crashes. HOS Compliance needs 3 relevant driver inspections with at least one of them showing an HOS violation; Vehicle Maintenance, Driver Fitness and HM Compliance need 5 relevant inspections with at least one violation among them. A relevant inspection is any inspection where the officer actually reviewed that area, clean ones included. There is also a recency filter: in several BASICs you are dropped from the ranking if no violation was recorded in the past 12 months and none appeared on your latest relevant inspection.

Why does a small carrier's score swing so hard from one inspection?

The answer is arithmetic, not bad luck. Four mechanisms stack:

1. The denominator is tiny. In Unsafe Driving the divisor is your average power units. A recent severity-8 violation carries a time weight of 3, entering the numerator as 24. Across 2 power units that is 12. Across 200 it is 0.12. Same event, two orders of magnitude apart.
2. You sit in Group 1, where one event is a large share of the record. With 3 inspections carrying Unsafe Driving violations you are at the floor of the combination range 3 to 8. A fourth changes your event count by a third. A Group 5 carrier with 150+ absorbs it invisibly.
3. Time weight decay hits harder. When your whole record is two violations, one dropping from weight 3 to 2 at six months rewrites your measure. With 200 violations, decay is a smooth curve.
4. Crossing a group boundary re-ranks you. Go from 8 to 9 qualifying inspections in combination Unsafe Driving and you leave Group 1 for Group 2, compared against a different population. Your percentile can move sharply while your own measure barely changed.

FMCSA has effectively conceded this. Its approved SMS update replaces safety event groups with proportionate percentiles, using a carrier's exact number of inspections and crashes instead of group cutoffs, so that results stay stable and there is no longer a substantial percentile increase without a corresponding measure increase. See the Prioritization Preview.

Why can't you see your own percentile publicly?

Because of the FAST Act of 2015. Property carriers' BASIC percentiles and alert symbols are not displayed to the public: the public page shows inspections, violations, crashes, out-of-service data, investigation results and your raw BASIC measures, but no percentile column and no alert symbol. FMCSA notes the measures are generated directly from safety data and are not a relative comparison to other carriers. Crash Indicator and HM Compliance are marked not publicly available regardless.

The carrier itself sees the full numbers by logging in at ai.fmcsa.dot.gov/sms with its DOT PIN: step by step in how to check your CSA score for free. Drivers are separate: there is no driver CSA score, and a driver's own history lives in the PSP report, covered in CSA scores for drivers and PSP.

What does FMCSA itself say about judging carriers by these numbers?

The federal position. FMCSA prints this on the SMS site: readers should not draw conclusions about a carrier's overall safety condition simply based on the data displayed in the system, and unless a carrier has received an UNSATISFACTORY safety rating under 49 CFR Part 385 or has otherwise been ordered to discontinue operations, it is authorized to operate on the nation's roadways. SMS data is explicitly not a federal safety rating under 49 U.S.C. 31144, a distinction untangled in safety rating vs CSA score.

The market reality. Brokers, shippers and underwriters read what is public and ask carriers for what is not. FMCSA neither requires nor endorses this, and makes no claim either way about how percentiles relate to insurance pricing or load access, but pretending the market ignores the data would be dishonest. How it plays out is in how brokers and shippers check CSA.

A practical target to replace "a good score"

• Zero percentiles at or above threshold. That is the only binary FMCSA acts on. One BASIC at 66% matters more than four sitting at 50%.
• Grow the denominator honestly. In the five inspection-normalized BASICs a clean inspection is not neutral, it dilutes your measure.
• Check every snapshot and dispute errors via a Request for Data Review at dataqs.fmcsa.dot.gov. Bad data works against you for up to 24 months.

FAQ

What is a good CSA score?

There is no single one. Each BASIC is scored separately from 0 to 100 with higher being worse, and each has its own threshold. Below threshold in every BASIC is the working definition of good.

What is a good CSA score for a trucking company?

For a general property carrier: under 65% in Unsafe Driving, Crash Indicator and HOS Compliance, and under 80% in the other four BASICs.

Is a CSA score of 50 good?

For a general carrier yes: below every threshold, meaning half your peer group has a worse measure. For a passenger carrier, 50% is exactly at the threshold in three BASICs.

What CSA score is too high?

Anything at or above the threshold for that BASIC. At, not just above. For general carriers, 65 in the strict three and 80 in the rest.

Why is my score worse than a carrier with the same violations?

It is a ranking divided by your size. Unsafe Driving and Crash Indicator divide by average power units, the other five by the time weight of your relevant inspections, and you are ranked only inside your own safety event group.

Why did my score jump with no new violations?

Time weights decay at 6 and 12 months, events drop off at 24 months, your safety event group can change, and every other carrier's data moves each monthly snapshot.

Can I see my exact percentile?

You can, the public cannot. Since the FAST Act of 2015 property carriers' percentiles are hidden publicly. Log in to SMS with your DOT PIN. Crash Indicator and HM Compliance are not publicly available at all.

Do brokers and insurers use CSA scores?

Many look at the data in practice. FMCSA does not endorse that use and states readers should not draw conclusions about a carrier's overall safety condition from these numbers alone.

Related articles

• CSA Scores Explained: BASICs, Thresholds, Methodology
• How to Check Your CSA Score for Free
• CSA Scores for Drivers: What PSP Really Shows
• Safety Rating vs CSA Score: Not the Same Thing
• How Brokers and Shippers Actually Check CSA
• HOS Violations and What They Cost in CSA
• When Do CSA Violations Fall Off Your Record?