595,851 Federal Sentences Exposed: What Your Judge Is Really Doing [2026]

You are sitting in a courtroom waiting for sentencing. You have done everything right. Character letters. Treatment enrollment. A mitigation package your attorney submitted three days early. You took responsibility. You showed up clean and employed and remorseful.

The judge reads from the presentence report. Opens a folder. Looks up.

And then gives you a sentence that is 40% above the guidelines range.

Your co-defendant, same charges, same case, different judge one courtroom over, got 30% below the range.

You will spend the next several years wondering what happened. Here is what happened: you drew the wrong judge. And nobody told you that mattered.

The Dataset Nobody Talks About

The United States Sentencing Commission has collected data on every federal sentence handed down since the early 2000s. Researchers at the QSIDE Institute compiled this into a publicly available dataset called JUSTFAIR, 595,851 individual sentencing records spanning FY2001 through FY2023.

That is not a sample. That is nearly every federal sentence imposed across 22 fiscal years, every district, every offense category.

And the data tells a story that defense attorneys know intuitively but rarely say out loud: which judge you draw matters as much as what you did.

Judges Have Patterns, The Data Proves It

Federal Sentencing Guidelines are advisory. After the Supreme Court's 2005 ruling, judges calculate the guidelines range but can sentence above or below it based on statutory factors. The result is enormous variation.

How enormous? Across the JUSTFAIR dataset, individual judges show downward departure rates ranging from under 10% to over 60%. Same courthouse. Same charge categories. Wildly different outcomes.

Here is what that means in plain English: if you are a defendant facing sentencing in a federal courthouse, the judge assigned to your case is one of the single biggest variables determining your outcome. Not the only variable. Not more important than the facts. But a variable that most defendants never examine.

Your guidelines range is a number. Your judge's history with that number is the context that tells you what the number actually means.

What the Data Actually Shows

The JUSTFAIR dataset tracks several key dimensions across those 595,851 records:

Departure rates vary wildly. The national average for below-range sentences has shifted over the years, but individual judges deviate from their own district's average by 20, 30, even 40 percentage points. The guidelines range your attorney calculated is a starting point, not a destination. The judge's historical relationship with that range is the map.

Appointing party correlates with sentencing severity. This is not a political statement. It is a statistical pattern visible across hundreds of thousands of records. Judges appointed by different presidents, across different eras, show measurable differences in sentencing patterns. Not deterministic, plenty of individual exceptions, but the aggregate trend is real and documented.

District culture matters. Federal courts in different parts of the country have different sentencing cultures. A defendant in the Southern District of New York faces a different statistical landscape than a defendant in the Western District of Texas, even for identical offenses. The USSC data shows persistent regional variation that no individual attorney's courtroom experience can fully capture.

Offense category changes the picture. A judge who departs downward frequently for drug offenses may rarely do so for fraud. Aggregate departure rates without offense-type filtering are misleading. The data supports offense-specific analysis, and most defendants never see that breakdown.

Why This Matters for Your Case

You are not a statistic. Your case has its own facts, its own circumstances, its own mitigating factors. Nobody is saying a departure rate number determines your sentence.

But here is what 595,851 records do tell you:

Your mitigation strategy should be calibrated to your judge, not to a generic template. A judge who almost never departs downward requires a different approach than one who does so regularly. The character letters, the treatment enrollment, the allocution, the emphasis shifts depending on who is listening.

Your plea negotiation changes meaning. A prosecutor's offer of 60 months means something different when your judge's average sentence for the same offense is 48 months than when it is 72. The plea discount or premium is invisible without context. The data provides the context.

The bench vs. jury decision has data behind it. USSC records track disposition method. Some judges sentence significantly differently after bench trials versus jury trials. That information exists. Most defendants make the bench-versus-jury choice without it.

Your attorney's courtroom instinct is valuable, but it is anecdotal. Even an attorney who has appeared before your judge 50 times has a sample size of 50. The dataset has hundreds or thousands of sentences from the same judge. Pattern recognition at that scale reveals things that courtroom experience alone cannot.

What You Can Do With This Information

We are not attorneys. We do not give legal advice. We do not tell you what sentence to expect or how to plead. That is between you and your attorney.

But we believe in one thing absolutely: the defendant should never be the least informed person in the courtroom.

The judge knows their own patterns. The prosecutor knows the judge's tendencies, they see them every week. Your attorney probably has a sense of the judge, built from however many cases they have handled in that courtroom.

You are the only person in that room who has never seen the data.

That is the information gap. And it is fixable.

Start free: The Federal Sentencing Calculator estimates your guidelines range using the same methodology federal probation officers use. It runs on JUSTFAIR data and takes about two minutes.

Go deeper: The Judge Report Card pulls your specific judge's sentencing history, departure rates by offense type, how they compare to the district and national averages, and patterns that are invisible from a single courtroom appearance. It is generated in under 60 seconds.

---

595,851 sentences. 22 years of data. Your judge has a pattern. You should know what it is before you walk into that courtroom.

Data referenced in this article comes from the JUSTFAIR dataset (QSIDE Institute), which aggregates U.S. Sentencing Commission data. This article is for informational purposes only and does not constitute legal advice.