Geographic Sentencing Disparity by District, FY2024

The highest above-benchmark districts

Among districts with at least 500 cases, Northern Alabama recorded the largest positive disparity in fiscal year 2024 at 60.4 percent above its benchmark across 1,113 cases, followed by Southern Texas at 55.1 percent over 1,112 cases. U.S. Sentencing Commission Individual Offender Datafiles, FY2024. Middle Alabama, Kansas, Northern Texas, and Idaho rounded out the top of the above-benchmark list, all sentencing at least a quarter higher than expected given their offense composition. Alabama appears twice in the top three, a clustering that suggests a regional pattern rather than a single outlier court.

A positive disparity does not mean a district is acting improperly. It means that, after accounting for the kinds of cases it sees, its imposed sentences ran higher than the national pattern would predict. U.S. Sentencing Commission Sourcebook of Federal Sentencing Statistics, 2024. That can stem from local charging practices, the specific firearms and fraud profile of a docket, or how often judges in the district impose above-range or upper-range sentences. The disparity figure flags where to look, not why the gap exists.

The lowest below-benchmark districts

At the other extreme, Guam sentenced 36.0 percent below its benchmark over 637 cases, the largest negative disparity, followed by New Mexico at 29.0 percent below across 2,621 cases. U.S. Sentencing Commission Individual Offender Datafiles, FY2024. Rhode Island, South Dakota, Maine, and Western Missouri also sat well below benchmark. Several of these districts run heavy immigration or lower-level drug dockets where fast-track programs and government-sponsored reductions push sentences below the otherwise-expected level, which is the mechanical driver behind much of the negative disparity.

The span between the extremes is wide. Northern Alabama at 60.4 percent above and Guam at 36.0 percent below sit nearly a hundred percentage points apart on this benchmark. U.S. Sentencing Commission Sourcebook of Federal Sentencing Statistics, 2024. That distance is exactly the kind of geographic variation the guidelines were designed to compress, and its persistence is a recurring theme in the Commission's own annual analysis.

How to read a disparity figure

Disparity here is a relative measure: it compares a district's sentences to what its offense mix would predict, not to a flat national average. A district handling unusually serious cases is not automatically high-disparity, because its benchmark already reflects that severity. U.S. Sentencing Commission Sourcebook of Federal Sentencing Statistics, 2024. The 500-case floor used here filters out small districts whose figures would swing on a handful of cases. Readers can examine any district's full record, including caseload and average term, on its individual district page, and the circuit-level view is in the companion circuit analysis.

Why disparity is the hardest measure to get right

Of every figure on this site, disparity is the one most easily misread, which is precisely why it deserves the most careful framing. A raw comparison of average sentences would label every border district as lenient and every fraud-heavy district as harsh, telling the reader nothing they could not have guessed from the offense mix alone. U.S. Sentencing Commission Individual Offender Datafiles, FY2024. The benchmark approach used here strips that predictable variation away and leaves only the residual, the part of a district's sentencing that its caseload does not explain. That residual is where the question of unwarranted geographic disparity actually lives, and it is the question the federal guidelines were written to address. A large residual does not condemn a court, and a small one does not vindicate it; both simply mark where the imposed pattern departs from the expected one. Presenting that residual honestly, with its caveats attached rather than buried, is the difference between a number that informs and a number that misleads, and it is the standard this analysis holds itself to.