Blog
Should I Take My Federal Case to Trial?
The Cost of Going to Trial
The federal trial is a wager with asymmetric consequences, and the terms are set before you enter the courtroom. In the federal system, the overwhelming majority of cases conclude with a guilty plea. Not because every defendant is guilty beyond question, but because the structure of federal sentencing converts the exercise of a constitutional right into a measurable cost. The question is whether your case possesses the characteristics that justify absorbing it.
Most people who contact our firm have already received a plea offer. The offer arrived, probably, in the form of a letter from the Assistant United States Attorney to defense counsel, outlining the charges the government is willing to let stand, the charges it will dismiss, and the sentencing recommendation it will support. The number at the bottom of that letter is not the sentence. It is the government’s opening position. The alternative is trial, and the alternative carries a price that the government does not always state in terms the defendant can parse.
What follows is an attempt to state them plainly.
Acceptance of Responsibility and the Sentencing Guidelines
Under Section 3E1.1 of the United States Sentencing Guidelines, a defendant who demonstrates acceptance of responsibility receives a reduction of two levels from the calculated offense level. If the defendant also assists the government by entering a timely guilty plea, permitting the government to avoid trial preparation and the court to schedule its calendar, the government may move for an additional one-level reduction. For defendants at offense level 16 or higher, this three-level reduction amounts to a sentence decrease of roughly thirty-five percent.
The reduction is, if we are being precise about it, not a reward for pleading guilty. The Guidelines commentary frames it as recognition of the defendant’s acceptance of personal responsibility for criminal conduct. In practice, the distinction between accepting responsibility and pleading guilty has collapsed. A defendant who proceeds to trial and is convicted will forfeit this reduction in nearly every case. The commentary states the adjustment “is not intended to apply to a defendant who puts the government to its burden of proof at trial by denying the essential factual elements of guilt, is convicted, and only then admits guilt.”
The National Association of Criminal Defense Lawyers examined federal sentencing data and found that trial sentences were roughly three times longer than plea sentences for comparable conduct. In drug trafficking cases, the disparity was steepest: sentences after trial averaged nearly three times the length of sentences after a plea.
The trial penalty exists in the background of every federal case, accumulating weight before the defendant has had the opportunity to assess it. The Supreme Court’s decision in United States v. Booker rendered the Guidelines advisory. A sentencing judge must consider the Guidelines range but retains discretion to depart from it under the factors set out in 18 U.S.C. § 3553(a). That discretion means the calculus is less mechanical than the raw numbers suggest, though I would not overstate how often judges exercise it in ways that neutralize the penalty. Some do. Many do not. The jurisdiction matters, and the individual judge matters more than the jurisdiction.
A three-level reduction sounds abstract until you calculate it against a Guidelines range. For a defendant at offense level 29 with a criminal history category of I, the difference between the range with acceptance and the range without it is the difference between roughly five years and roughly eight. That is not an abstraction.
There is the further question of what occurs at sentencing after a trial the defendant loses. The judge who presided has now observed the defendant’s conduct, the full scope of the government’s evidence, the credibility of each witness under cross-examination. Most federal defendants who plead guilty are sentenced by a judge who has reviewed a presentence report and perhaps heard a brief allocution. The judge at a post-trial sentencing has seen everything. Whether judges impose a penalty for the exercise of trial rights as a conscious matter is debatable, and the debate has persisted for decades without resolution. What is not debatable is the informational difference. The judge knows more after a trial. What the judge comes to know does not always favor the defense.
The Government’s Case and What It Conceals
Before the trial decision becomes ripe, there is discovery. In federal practice, the government’s obligations under Brady v. Maryland and Federal Rule of Criminal Procedure 16 require disclosure of material favorable to the defense, as well as specific categories of evidence. The government also produces, typically on the eve of trial, the materials required under the Jencks Act: prior statements of witnesses who will testify.
Discovery is where the decision begins to take shape. The strength of the government’s evidence, the reliability of cooperating witnesses (one of whom, in a case we reviewed last year, had provided three materially different accounts of the same transaction to three different agencies before settling on the version the government preferred), the existence of recorded communications, the forensic integrity of physical evidence. These are documents. They tell you whether the government can prove its case beyond a reasonable doubt.
Federal prosecutors do not indict cases they expect to lose. The conviction rate at trial hovers near eighty-five percent for jury verdicts. The government screens before charging, and the cases that survive tend to be strong. Our approach to discovery begins with the assumption that the government believes it can prove every element. We then test that belief against the actual evidence, piece by piece, looking for the discrepancy between what the indictment alleges and what the exhibits establish.
In something like a quarter of the cases we review (a figure I offer from practice, not from a dataset), discovery reveals a fracture the government has not addressed. A cooperating witness whose prior statements contain inconsistencies. A forensic methodology under challenge in other circuits. A gap in the chain of possession that renders a key exhibit vulnerable. These fractures do not guarantee an acquittal. They change the calculus.
Whether a fracture is sufficient to justify the cost of trial requires an assessment that considers the nature of the weakness, the judge, the probable jury pool, the defendant’s history, and the distance between the plea offer and the probable post-trial sentence. It is possible to identify a genuine problem in the government’s proof and still conclude that trial is inadvisable. This is the portion of the analysis that resists generalization.
When Trial Is the Correct Decision
Trial is correct when the plea offer is no offer at all. In cases involving mandatory minimum sentences where the government has declined to file a motion under Section 5K1.1 for substantial assistance, the defendant may face the same statutory floor whether pleading guilty or proceeding to trial. If the floor does not move, the cost of trial diminishes.
Trial is correct when the defendant is innocent. An innocent defendant who pleads guilty to avoid the trial penalty has accepted a conviction and a criminal record that belong to someone else. We have sat across the table from people in this position: a business owner who signed documents a partner prepared, a driver who did not know what was in the vehicle, a name on a lease that became the basis for a conspiracy charge. You sign the plea agreement and then you discover what the plea agreement means. The conversation that follows is among the most difficult in criminal practice.
Trial is correct when the government’s case depends on a single cooperating witness whose credibility is impeachable, when suppression motions present a realistic prospect of excluding central evidence, or when the legal theory underlying the indictment is novel enough that a court might reject it. There are exceptions, though in practice they tend to confirm what the numbers already suggest.
Trial is occasionally correct for reasons that resist the framework of cost and benefit. Some clients arrive at the decision after absorbing every number, understanding every risk, and concluding that they will not plead guilty because the system must be required to prove its case. I respect that decision when it is informed. I have seen it produce acquittals that no practitioner would have predicted. Whether the court intended that outcome or merely failed to prevent it is a question worth asking.
What the Courtroom Requires
In 2019, before the current wave of sentencing reform discussions reached Congress, a drug trafficking defendant in the Southern District of West Virginia proceeded to trial on the advice of counsel who had never tried a federal case. The trial lasted four days. The conviction took the jury less than three hours. The sentence was more than twice what the plea offer had been. The case is not unusual. What made it visible was the district judge’s subsequent commentary on the state of plea bargaining in federal court, which generated attention that the defendant’s sentence did not.
A federal trial is a procedure governed by the Federal Rules of Evidence, the Federal Rules of Criminal Procedure, and the particular practices of the district. The way the Southern District of New York handles jury selection is not the way the Eastern District of Virginia does it, and counsel who has not appeared before the assigned judge operates at a disadvantage that preparation alone cannot fully address. I am less confident about this in districts where the bench turns over frequently, but the general principle holds.
The jury receives instructions that define the elements the government must prove. The verdict must be unanimous.
The cases that go to trial tend to fall into two categories: cases where defense counsel has identified something specific and articulable in the government’s proof, and cases where the defendant has concluded that the plea offer is unacceptable on its terms. The first category produces better results. Both categories produce acquittals, and both produce convictions that exceed what the plea offered.
The decision to proceed to trial in a federal case is a decision about evidence, about sentencing exposure, about the assigned judge and the jury pool, and about what you are prepared to accept if the verdict is not what you wanted. A consultation is where that assessment begins, and it assumes nothing.