Process

Plea Bargaining in Federal Court

Over 97 percent of federal cases end in guilty pleas. How plea bargaining works and how it shapes sentencing data.

Disclaimer: PlainSentencing is for educational and research purposes only. Nothing on this site constitutes legal advice.

The Dominance of Guilty Pleas

Over 97 percent of federal criminal cases are resolved through guilty pleas rather than trial. This statistic — consistent year after year in USSC data — means that plea bargaining is the primary mechanism through which federal sentences are determined. Understanding how plea bargaining works is essential for interpreting the sentencing data on PlainSentencing.

How Federal Plea Bargaining Works

Federal plea bargaining typically takes one of three forms recognized by the Federal Rules of Criminal Procedure:

  • Charge bargaining — The government agrees to dismiss certain charges in exchange for a guilty plea to others. This directly affects the guideline calculation by changing the offense of conviction.
  • Sentence bargaining (Rule 11(c)(1)(C)) — The parties agree on a specific sentence or sentencing range, which the judge may accept or reject. If the judge rejects it, the defendant can withdraw the plea.
  • Recommendation bargaining — The government agrees to recommend a particular sentence or not oppose a defense request. The judge is not bound by the recommendation but typically gives it weight.

The Trial Penalty

Defendants who go to trial and are convicted face significantly longer sentences on average than those who plead guilty to the same charges. This difference — sometimes called the trial penalty — reflects several factors: defendants who plead guilty receive the acceptance of responsibility reduction (2-3 guideline levels), cooperative defendants may receive substantial assistance departures, and prosecutors may offer favorable plea deals to avoid the cost and uncertainty of trial. The trial penalty is one of the most powerful incentives driving the high guilty plea rate.

How Plea Bargaining Shapes Sentencing Data

Because plea bargaining determines both the charges and often the sentencing recommendations for the vast majority of cases, the data on PlainSentencing reflects negotiated outcomes more than pure judicial sentencing decisions. Average sentence lengths, guideline compliance rates, and disparity scores are all influenced by prosecutorial charging and plea practices — which vary by district and by U.S. Attorney office.

This context is important for interpreting geographic disparities. A district with lower average sentences may not have more lenient judges — it may have prosecutors who more frequently offer favorable plea agreements or file substantial assistance motions. The data cannot fully separate judicial discretion from prosecutorial discretion.

The Role of Cooperation

Cooperation with the government — providing information or testimony that helps investigate or prosecute other cases — is the most powerful sentence-reduction tool in the federal system. Defendants who provide substantial assistance can receive sentences below the guideline range and even below mandatory minimums. The government controls this process by deciding whether to file a 5K1.1 motion, giving prosecutors significant leverage during plea negotiations.

Explore the Patterns

PlainSentencing tracks plea rates, guideline compliance, and departure patterns across all federal districts. Browse district profiles to see how these patterns differ by geography, or explore offense categories to see how plea and sentencing dynamics vary by crime type.

The Transparency Gap

One significant limitation of sentencing data — including the data on PlainSentencing — is that plea negotiations happen largely behind closed doors. The final sentence reflects the negotiated outcome, but the data cannot reveal what charges were originally considered, what deals were offered and rejected, or what factors drove the final agreement. This means that sentencing data captures outcomes but not the full process that produced them.

Despite this limitation, the patterns visible in PlainSentencing data — geographic variation in average sentences, differences in guideline compliance rates, varying cooperation rates — all provide meaningful insight into how the federal sentencing system operates in practice. The data may not tell the complete story, but it tells an important one.

Why This Matters for Research and Journalism

Federal sentencing data is one of the most important public-records corpora produced by the U.S. government. Researchers, journalists, defense attorneys, and policy analysts rely on it to evaluate the fairness, consistency, and proportionality of the federal criminal justice system. PlainSentencing exists to make this dataset navigable for non-specialists without flattening the complexity that makes it meaningful.

Every district profile, ranking, and explainer on this site is derived from the U.S. Sentencing Commission Individual Offender Datafiles, which we recompute from the raw FY2015 through FY2024 records. The numbers presented here are reproducible from the source data, and the methodology that produced each chart and table is documented on our methodology page. Corrections, refinements, and questions are welcome at the contact address listed there.

For practitioners considering using this data in court filings, scholarly research, or journalism, please cite the underlying USSC datasets rather than PlainSentencing — we are a presentation layer, not the primary source. Our editorial role is to translate technical fields, explain methodological caveats, and surface comparisons that highlight meaningful patterns in the federal sentencing system.

Frequently Asked Questions

Where does PlainSentencing get its data?

All data comes from the United States Sentencing Commission (USSC) public-use datafiles. The USSC collects data on every federal offender sentenced in U.S. district courts and publishes annual datafiles. PlainSentencing aggregates FY2015 through FY2024, covering over 660,000 sentencing records.

What does the disparity score measure?

The disparity score measures how much a district average sentence deviates from the national average for the same offense types. Higher scores indicate greater deviation. It controls for offense mix but not for case-specific factors like cooperation or role in offense.

Is PlainSentencing legal advice?

No. PlainSentencing is for educational and research purposes only. Sentencing outcomes depend on case-specific factors that aggregate statistics cannot capture. Consult a qualified federal defense attorney for guidance on any specific case.

Worked example: a typical federal plea agreement

A defendant indicted on three counts (drug conspiracy, firearm possession, money laundering) faces a combined statutory maximum of 50 years. Through plea negotiation, the defendant agrees to plead guilty to one count of drug conspiracy in exchange for dismissal of the firearm and laundering counts. The government recommends a 3-level acceptance-of-responsibility reduction under USSG §3E1.1. The base offense level drops from 32 to 29; the guideline range falls from 121-151 months to 87-108 months. The defendant also agrees to a 5K1.1 cooperation provision: testimony against two co-defendants in exchange for a further 25% downward departure. Final sentence: approximately 65 months — substantially below the 121-month exposure under conviction at trial.

Plea rates by offense category

Offense categoryPlea rateAvg. reduction
Drug trafficking97%32%
Firearms96%22%
Fraud and embezzlement98%28%
Immigration99%15%
Child exploitation94%18%
White-collar97%26%
Violent crime92%20%
National average97%25%

When plea negotiations fail

About 3% of federal cases proceed to trial. Of those, the conviction rate is approximately 88% — substantially higher than in state systems, partly because federal prosecutors are heavily selective about which cases they indict and partly because federal trials see fewer evidentiary suppressions. Trial sentences average 40-60% longer than equivalent plea sentences for the same offense conduct, primarily because trial defendants forfeit acceptance-of-responsibility reductions and rarely earn cooperation departures. The "trial penalty" is the single most-debated structural feature of federal sentencing — defenders argue it coerces pleas; prosecutors argue it reflects the additional resources expended at trial and the absence of cooperation. The empirical reality: a defendant who goes to trial and is convicted faces, on average, an additional 4-7 years of incarceration relative to the same defendant who would have pleaded.