Guide to White-Collar Criminal Defense: Federal Sentencing Explorer

Federal Sentencing Explorer

Free interactive access to 805,875 federal sentences — FY2014 through FY2025.

Launch the Federal Sentencing Explorer →

Every federal sentencing starts with a guideline range. It rarely ends there. In fiscal year 2025, federal judges sentenced within the range just 51.5% of the time. The Federal Sentencing Explorer, built by Dynamis LLP from the U.S. Sentencing Commission's public datafiles, shows what actually happened to defendants comparable to you or your client — by offense, guideline level, criminal history, and district.

What the Explorer does

The Explorer puts twelve years of Sentencing Commission data — every federal defendant sentenced from FY2014 through FY2025 — behind five filters:

Offense type. Thirty categories, from fraud and drug trafficking to firearms, tax, and money laundering. Economic crimes break down further into 21 subtypes: securities fraud, healthcare fraud, government benefits, embezzlement, identity theft, and more.

Final offense level and criminal history category. Match the guideline posture of your case — the offense level and CHC the court actually applied, as reported by the Commission.

District. All 94 federal districts. Compare your district's record against the national picture. Median fraud sentences vary several-fold between districts applying the same guidelines.

Cooperation status. Substantial-assistance (§5K1.1) cases are coded separately. One checkbox removes cooperators from your comparison pool — the difference between a benchmark a prosecutor can dismiss and one that holds up.

For any combination, the Explorer returns the number of comparable cases, the median and mean sentence, the full percentile spread, the probation rate, and how often judges imposed below-range sentences on their own initiative — plus a one-click citable summary formatted for a sentencing memorandum.

Who it is for

Defendants and their families. The most common question we hear is the one no lawyer can answer with certainty: what is going to happen to me? The Explorer will not predict your sentence. It will show you the honest range of outcomes for people in comparable positions — which is a far better starting point than fear.

Defense lawyers. Judges respond to data about their own districts. A memorandum that shows the court what comparable defendants received — with cooperators filtered out and the guideline posture matched — makes the § 3553(a) argument in the court's own language. The citable-summary feature generates the sentence for you, sourced to the Commission's datafiles.

Researchers and journalists. The underlying data is public, but the Commission's files are 20,000-variable research datasets that require statistical software. The Explorer makes the core sentencing questions answerable in a browser, free, with the methodology documented on the page.

How to use it: an example

Suppose your client faces sentencing for securities fraud, final offense level 22, Criminal History Category I, in the Southern District of New York. In the Explorer: set Offense Type to Fraud/Theft/Embezzlement, Economic Crime Subtype to Securities/Investment, offense level 22 to 22, CHC I, district S.D.N.Y. — and uncheck §5K1.1 to exclude cooperators. The tool returns the comparable cases and their outcomes, and the copyable summary reads like this:

Per U.S. Sentencing Commission Individual Offender Datafiles, FY2014–FY2025, of the defendants sentenced in the S.D.N.Y. for economic crimes of the Securities/Investment type with final offense level 22 and Criminal History Category I (excluding §5K1.1 substantial-assistance cases), the median sentence was X months...

Widen the filters — more years, neighboring offense levels, national scope — until the pool is large enough to be meaningful, and narrow them when precision matters more than sample size.

Methodology and limitations

The Explorer runs on the Sentencing Commission's public Individual Offender Datafiles (FY2014–FY2025) and Economic Crime datafiles, harmonized across the Commission's 2018 and 2020 coding changes. Sentences include prison plus alternatives, with probation counted as zero months and sentences of 470 months or more (including life) capped at 470. The public data contains no judge identifiers — the district is the finest available unit. Full methodology notes appear at the bottom of the tool itself.

Two honest caveats. First, historical statistics describe patterns; they do not predict any individual outcome. Second, comparator pools blend cases that differ in ways the data cannot capture — loss amounts, victims, role, acceptance. The numbers frame the argument. They are not the argument.

Launch the Federal Sentencing Explorer →

Facing a federal sentencing?

Data is where preparation starts, not where it ends. Dynamis LLP's white-collar defense team — led by former federal prosecutors — has represented individuals and companies at sentencing in districts across the country. Contact us for a confidential consultation.


This page and the Federal Sentencing Explorer are provided for informational purposes only and do not constitute legal advice. Use of the tool does not create an attorney-client relationship with Dynamis LLP. Attorney advertising. Prior results do not guarantee a similar outcome.

Further Resources on White-Collar Criminal Defense