by Doug Ankney

“I consider trial by jury as the only anchor ever yet imagined by man, by which a government can be held to the principles of its constitution.” —Thomas Jefferson (Personal letter to Thomas Paine, dated July 11, 1789)

The American criminal justice system is built on a constitutional promise: the right to a trial by jury, a protection John Adams called “the heart and lungs of liberty.” This public, adversarial process was designed to be the primary defense against government overreach, ensuring that a defendant’s fate would be decided by a jury of their peers. Today, this promise is a statistical outlier. In the federal system, over 97% of all convictions are obtained not through trial but through a guilty plea, a figure mirrored in state courts. The system has transformed from one designed for public trials into one dominated by private negotiations conducted behind closed doors. As Justice Anthony Kennedy observed, plea bargaining today “is not some adjunct to the criminal justice system; it IS the criminal justice system.”

This transformation is not voluntary; it is coerced. Prosecutors – armed by legislatures with mandatory minimum sentences, charge-­stacking provisions, and sentencing enhancements – wield the power to make the risk of trial so catastrophically high that defendants, guilty and innocent alike, are forced to plead guilty. Legal scholars have documented how these tools are deployed not for justice but to gain “plea-­bargaining leverage.” The result is a system that prioritizes efficiency over constitutional rights, expedience over truth, and coercion over fairness.

The human cost of this coercion is not abstract. When Georgia prison officials executed Ray Cromartie in 2019, he still maintained his innocence in the 1994 murder for which he was convicted. He had by then made numerous requests for DNA testing of crime scene evidence, believing it would exonerate him. But every request was denied. That was punishment for rejecting a plea deal offered before his 1997 trial that would have made him eligible for parole after just seven years. Instead, Cromartie’s faith in the criminal justice system meant he spent over 22 years on death row before the state took his life.

Cromartie’s case is but one example of what the National Association of Criminal Defense Lawyers (“NACDL”) calls the “trial penalty,” the substantial and often devastating difference between the sentence offered in a plea and the sentence imposed after trial. This practice occurs daily in courthouses across America, where prosecutors threaten harsh penalties to those who dare exercise their constitutional rights. How did going to trial become so dangerous? Why do prosecutors hold such vast, unchecked power? And what can be done to restore balance?

This article addresses these questions in three parts. Part 1 traces the history of jury trials in America and the rise of plea bargaining to its current dominance. Part 2 examines the trial penalty’s devastating impact on individuals, communities, and the criminal justice system’s legitimacy. Part 3 explores reform efforts underway and recommendations from judges, legal scholars, and advocacy organizations to rein in prosecutorial power and meaningfully restore the constitutional right to trial.

Part 1: Justice Gives Way to (Coerced) Efficiency

Jury trials are often portrayed as foundational to democracy and the backbone of the American criminal justice system. And rightly so. At America’s founding, and for a time thereafter, all criminal cases were decided by juries. Enshrined in both Article III and the Sixth Amendment to the United States Constitution, the right to a jury trial is deeply embedded in the American democratic ethos. Juries were the norm in colonial America, ensuring democratic representation and thwarting the threat of despotic power in the criminal justice system.

The reverence for jury trials ran deep in American culture. Vikrant Reddy, Senior Fellow at Stand Together and the Charles Koch Institute, recounts a striking mid-­19th century example: when a group of pioneers traveling west discovered one of their party had been murdered, they halted their entire journey to conduct a trial, even though none of them was an attorney or judge. They were so concerned about impartiality that they waited for another wagon train to arrive, so they could recruit a judge and jurors. This was the jury trial’s place in the American psyche: not a procedural formality but a sacred obligation.

The Emergence and Spread of Guilty Pleas

By the late 19th century, this commitment to jury trials began to erode. Guilty pleas emerged as an alternative means of resolving cases, beginning in major urban jurisdictions and spreading rapidly. By the early 20th century, guilty pleas were commonplace. Several factors drove this expansion: escalating case­load pressures, longer and more complex trials with increasingly technical procedural rules, changes in policing and evidence collection, and the shift from appointed prosecutors to elected ones, making high conviction rates a political prize worth pursuing.

Despite this expansion, plea bargaining initially met judicial resistance. Because negotiations occurred in secret – off the record, in prosecutors’ offices or courthouse corridors – courts viewed the practice with suspicion. Some judges issued reasoned opinions highlighting legitimate concerns and seeking to check or even end the practice. In 1957, Judge Rives of the U.S. Court of Appeals for the Fifth Circuit wrote, “justice and liberty are not the subjects of bargaining and barter.” Shelton v. United States, 246 F.2d 571 (5th Cir. 1957) (Rives, J., dissenting). Following his lead, the Sixth Circuit declared, “[i]t is clear … that a plea of guilty induced by a promise of lenient treatment is an involuntary plea and hence void.” Scott v. United States, 349 F.2d 641 (6th Cir. 1965).

Until the 1970s, the U.S. Supreme Court shared this distrust of coerced plea bargains. Beginning with Walker v. Johnston, 312 U.S. 275 (1941), the Court ruled that a guilty plea induced by a prosecutor’s threat to seek a higher sentence was unconstitutional. This principle held firm. By 1968, not a single guilty plea had survived on appeal if the Court found evidence it was induced by the sorts of threats or promises now commonplace in plea bargaining.

But in 1970, the Supreme Court reversed course entirely. In Brady v. United States, 397 U.S. 742 (1970), the Court ruled that it was constitutional for prosecutors to offer inducements to obtain guilty pleas, including inducements that were, in effect, threats to seek higher sentences after trial. The Supreme Court reinforced this holding in Parker v. North Carolina, 397 U.S. 790 (1970), stating “an otherwise valid plea is not involuntary because induced by a defendant’s desire to limit the possible maximum penalty to less than that authorized if there is a jury trial.”

By 1971, the Supreme Court had fully embraced plea bargaining, characterizing it as “an essential component of the administration of justice” that “is to be encouraged.” Santobello v. New York, 404 U.S. 257 (1971). By the end of that decade, the Supreme Court had no problem at all with prosecutors coercing guilty pleas, as evidenced in Bordenkircher v. Hayes, 434 U.S. 357 (1978).

The facts of Bordenkircher are troubling. A prosecutor initially offered the defendant five years’ incarceration in a check forgery case. The prosecutor then threatened to seek additional charges carrying up to life in prison if the defendant rejected the plea offer. The defendant rejected the offer, so the prosecutor amended the charges. After conviction at trial, the defendant was sentenced to life in prison. The Supreme Court upheld these actions, ruling that the prosecutor acted constitutionally because the amended charges were properly chargeable and the defendant had been advised by competent counsel. The Court explained it “has necessarily accepted as constitutionally legitimate the simple reality that the prosecutor’s interest at the bargaining table is to persuade the defendant to forego his right to plead not guilty.” The trial penalty had received the Supreme Court’s official blessing.

The Sentencing Reform Act: Empowering Prosecutors

Following close on the Supreme Court’s reversal of jurisprudence on plea bargains came the Sentencing Reform Act of 1984 (“SRA”). Prior to the SRA, trial judges were limited in the sentence they may impose only by the Constitution or specific statutes. But that led to highly disparate outcomes for defendants whose cases were hard to distinguish except for their judge or the court. Consequently, Congress enacted the SRA to limit judicial discretion.

The SRA established a commission to develop the U.S. Sentencing Guidelines (“Guidelines”) that calculate a “sentencing range” based primarily on characteristics of the defendant and of the crime. It was mandatory in federal courts that judges impose sentences within the calculated sentencing range. Absent the few circumstances designated in the Guidelines, federal judges no longer had the option to choose a sentence outside the range, so their opinion as to its fairness effectively became irrelevant.

The Guidelines were (and are) skewed in favor of the prosecution, allowing prosecutors to seek longer sentences based on conduct of the defendant that was neither charged nor litigated, and even based on conduct for which the defendant was acquitted until a 2024 amendment to Guidelines §1B1.3 (Relevant Conduct), effective November 1, 2024, prohibited the use of acquitted conduct to increase the Guidelines range (though courts may still consider it under statutory authority). And tellingly, the Guidelines offer sentence reductions to defendants who plead guilty quickly.

The SRA also provided for “mandatory minimum sentences” on a host of crimes. Mary Price, general counsel for Families Against Mandatory Minimums (“FAMM”) explains that “[m]andatory minimums and their impact on the sentencing guidelines put prosecutors in a very strong position when negotiating plea accords. Prosecutors control which charges to bring, whether to charge cases so as to trigger mandatory minimums upon conviction, whether and how to drive guidelines ranges by manipulating relevant conduct of drug quantity [in drug cases] or [financial] loss [in fraud-­related cases], and whether to insist on aggravators, such as obstruction of justice enhancements when a defendant testifies at trial. The government also has a great deal of control over which charges to combine or dismiss, which relevant conduct to ignore, and whether to notice [§] 851 enhancements. Presently, prosecutors are instructed to charge the most serious, readily provable offenses, and severity is assessed by the length of sentence that is triggered upon conviction.”

And the SRA permitted prosecutors to “stack” charges, i.e., charge multiple violations based on the same conduct of the defendant. These procedures enabled prosecutors to threaten extremely long sentences that judges no longer had discretion to adjust when “inducing” the accused to plead guilty. Or, to paraphrase The Godfather, prosecutors make defendants “offers they can’t refuse.”

In United States v. Booker, 543 U.S. 220 (2005), the Supreme Court finally struck down the provision of the SRA that made the Guidelines mandatory. Theoretically, this restored sentencing discretion to judges and made the Guidelines “advisory.” But in practice, the calculated sentencing ranges remain the presumptive “starting point and benchmark” for all federal cases, as the Supreme Court subsequently admitted in Gall v. United States, 552 U.S. 38 (2007). Indeed, Justice Sotomayor recently explained that “[i]n most cases, it is the range set by the Guidelines, not the minimum or maximum term of imprisonment set by statute, that specifies the number of years a defendant will spend in prison.” Beckles v. United States, 580 U.S. 256 (2017). And the data reveals that over 80% of sentences imposed in the U.S. District Courts are within the calculated Guidelines range.

Under the SRA, as codified at 18 U.S.C. § 3553(a)(2), Congress required that sentences for criminal conduct should: “reflect the seriousness of the offense, promote respect for the law, and provide just punishment; afford adequate deterrence to criminal behavior; protect the public from future crimes of the defendant; and provide the defendant with needed education, training, health care or other correctional treatment in the most effective manner.” Sentences are to be adjusted in light of a defendant’s history and the manner in which the crime was committed and to avoid “unwarranted disparity among similarly situated defendants,” Price noted. Courts are to impose “sentences that are sufficient but not greater than needed to meet these goals.”

The Justice Manual (formerly the U.S. Attorney’s Manual) explicitly forbids bringing more severe charges simply to coerce a plea: “Charges should not be filed simply to exert leverage to induce a plea nor should charges be abandoned to arrive at a plea bargain that does not reflect the seriousness of the defendant’s conduct.” Justice Manual, section 9-­27.400. But it is well known that prosecutors ignore the Justice Manual’s directive, Price said, quoting from a 2011 report from the U.S. Sentencing Commission titled Mandatory Minimum Penalties in the Federal Criminal Justice System, which revealed that, based on interviews of both prosecutors and defense attorneys, “charges carrying mandatory minimum sentences were routinely manipulated to threaten defendants who balked at pleading and to reward those who complied with prosecutors demands.”

The Department of Justice (DOJ) has, over the years, “sought new and increased mandatory minimums ‘not because the enhancements are inherently just or required for adequate deterrence, but precisely because higher sentences provide increased plea bargaining leverage,’” according toFrank O. Bowman III. “American Buffalo: Vanishing Acquittals and the Gradual Extinction of the Federal Criminal Trial Lawyer,” University of Pennsylvania Law Review Online 156 (2007): 226.

Another tactic employed by prosecutors is the practice of an “exploding” or “drop dead” plea offer. This is an offer of leniency that expires if not accepted almost immediately – sometimes in a matter of minutes or by the end of the day – and a warning that a more burdensome offer will replace it. This practice is viewed as deeply coercive because its primary purpose is to manufacture panic and prevent a rational, counseled decision. It effectively bypasses the Sixth Amendment right to effective counsel by making it impossible for a defense attorney, who may have just received the case, to adequately review evidence (which they likely have not received yet), investigate the facts, or advise their client on the actual risks of trial. The defendant is therefore forced to make an irreversible choice about their liberty under extreme duress, turning the constitutionally-­required “voluntary” nature of a plea into a high-­pressure, non-­negotiable demand. This tactic is especially effective at forcing pleas from the innocent, who are denied the very time needed for their defense to uncover exculpatory facts.

Global Influence and the Near-­Extinction of Trials

In the 2017 Fair Trials Report titled The Disappearing Trial – Towards a Rights Based Approach to Trial Waivers (“Report”) it was reported that 65 countries use some form of plea bargaining and that the U.S. “was often the inspiration for the adoption of the practices.” The Report also stated that plea bargaining occurs more frequently in the U.S., with “fewer safeguards and more coercive practices than in any other nation in the world.”

The effect of coercive plea bargaining, with judicial approval and congressional enabling, on both jury trials and bench trials has been their near extinction. The dramatic decline in trials has swept through both the federal and state judicial systems. In the federal criminal justice system, 8.2% of cases were resolved by a jury trial in 1962, dropping to less than 5% in 2002 and only 3.6% by 2013. In 2021, of the 63,725 cases resolved in U.S. District Courts, 58,516 (92%) ended in criminal convictions. And 57,631 (98%) of those convictions came via guilty pleas. As recently as fiscal year 2024, the U.S. Sentencing Commission reported that 97% of all federal convictions stemmed from guilty pleas, with trials occurring most often in individual rights cases (32%) and murder cases (22%).

The state systems reflect the same trend toward trial extinction. In those states that maintained accurate records of jury and bench trials, the data reveals that between 1976 and 2002, trials for felonies fell from 3.4% to 1.3%. In 2017, all 22 of those states reported that the percentage of jury trials in courts of general jurisdiction was below 3%, and in most states, it was below 2%.

Justice Anthony Kennedy observed, “[C]riminal justice today is for the most part a system of pleas, not of trials.” Lafler v. Cooper, 566 U.S. 156 (2012). And in Missouri v. Frye, 566 U.S. 134 (2012), Kennedy wrote “plea bargaining … is not some adjunct to the criminal justice system, it IS the criminal justice system.” (emphasis in original)

Part 2: Destruction of Justice and Lives

Part 1 traced how the trial penalty transformed America’s criminal justice system from one centered on public jury trials to one dominated by coerced pleas. Part 2 examines the devastating consequences of this transformation, not just statistics, but the human cost and the erosion of constitutional protections that once safeguarded liberty.

The volume of cases resolved by plea agreements has a corrosive effect on the integrity of the criminal justice system. Former President John Adams understood what was at stake: “Representative government and trial by jury are the heart and lungs of liberty. Without them we have no other fortification against being ridden like horses, fleeced like sheep, worked like cattle, and fed and clothed like swine and hounds.” When defendants plead guilty to avoid the trial penalty, they waive protections the Framers considered essential to preventing tyranny.

As the NACDL summarized, the jury trial protections lost in a plea-­dominated system include:

The right to be found guilty only by a jury of one’s peers, selected with defense input under restrictions preventing discrimination

The right to discover exculpatory and impeachment evidence material to the case

The right to confront and cross-­examine witnesses through live, adversarial testing

The right against adverse prosecutorial comments regarding the choice to remain silent

The right to be found guilty only by unanimous jury verdict based on proof beyond a reasonable doubt

The right to raise constitutional challenges to how the government acquired evidence

The right to appeal the conviction and underlying rulings

Collectively, these protections form a web of checks against government overreach. When 97% of defendants waive them to avoid the trial penalty, the Constitution’s promise of due process becomes an impotent formality.

The consequences of waiving these rights extend beyond individual cases. As Martín Antonio Sabelli, former President of the NACDL, explained: “Not only does the right to trial wither in the brave new world of modern coercive plea bargaining, modern plea agreements also routinely require the waiver of every right, liberty, and freedom that the Framers established in the Bill of Rights, undermining the letter and spirit of the Fourth, Fifth, Sixth, and Eighth Amendments.” Challenging police misconduct, requesting bail consistent with the presumption of innocence, reviewing evidence, investigating a defense – each of these constitutionally necessary practices is routinely waived in the face of prosecutorial threats. The result, as Sabelli notes, is “the assembly line of pleas without the ‘inconvenience’ and ‘delays’ associated with evidentiary hearings, investigation of facts, and trial.”

But the loss of these procedural protections does more than eliminate inconvenience for prosecutors. It creates an environment where misconduct and injustice can flourish unseen.

Hidden Misconduct: How Plea Bargains Conceal Injustice

Without trials, police and prosecutorial misconduct as well as investigative and forensic errors that would normally be uncovered during the pre-­trial phase and at trial remain hidden. In a system where approximately 97% of criminal cases are resolved through plea bargains, the government’s evidence and procedures escape the rigorous scrutiny of adversarial testing. Prosecutors and police can secure convictions based on flawed, tainted, or even fabricated evidence because defendants, facing the trial penalty’s catastrophic sentences, plead guilty rather than risk going to trial.

This dynamic conceals two critical categories of prosecutorial violations: Brady violations (failure to disclose exculpatory evidence, as mandated by Brady v. Maryland, 373 U.S. 83 (1963)) and Giglio violations (failure to disclose impeachment evidence such as witness deals, as mandated by Giglio v. United States, 405 U.S. 150 (1972)). When defendants plead guilty to avoid decades in prison, they forfeit the opportunity to test the prosecution’s case through cross-­examination, motions to suppress, or expert analysis. The result is a system that prioritizes efficiency over truth-­seeking, allowing grave injustices, including the wrongful conviction of innocents, to persist without consequence.

Prosecutors, knowing that the overwhelming majority of cases will not go to trial, have little incentive to thoroughly vet their evidence or disclose weaknesses. Plea negotiations often occur before full discovery, allowing critical information to remain hidden from the defense. This practice is sanctioned by the Supreme Court’s 2002 ruling in United States v. Ruiz, 536 U.S. 622 (2002), which ruled that prosecutors are not required to disclose impeachment evidence (evidence that could undermine a witness’s credibility) before a plea is finalized.

The case of former U.S. Senator Theodore Stevens (R-Alaska) vividly illustrates this dynamic: prosecutors withheld exculpatory witness notes and debriefings that directly contradicted their theory, leading to a jury conviction. However, before Stevens was sentenced, an FBI whistleblower exposed the misconduct. In an extraordinary move, the DOJ itself asked the court to set aside the verdict. The trial judge vacated the conviction and dismissed the underlying indictment with prejudice, citing the most egregious prosecutorial misconduct he had ever seen. This injustice was only discovered because Stevens was in the tiny fraction of defendants who went to trial. Had Stevens pleaded guilty, as 97% of federal defendants do, the exculpatory evidence would have never surfaced because plea agreements universally require defendants to waive their trial rights and limit their appeals – the very mechanisms that expose such violations.

This concealment is systemic, not exceptional. A 2013 NACDL report found that in plea-­dominated systems, Brady violations are rarely litigated because defendants, facing overwhelming pressure and relying on overworked counsel, accept deals without knowing the full extent of favorable evidence. The consequences are staggering. According to the National Registry of Exonerations (“NRE”), 23% of all exonerees had pleaded guilty to crimes they did not commit. These individuals spent years, sometimes decades, in prison based on untested allegations, separated from families and stigmatized for life, because they feared the alternative was worse.

The problem is compounded by legal doctrines that actively enable concealment. While the Supreme Court’s 1963 ruling in Brady v. Maryland requires prosecutors to disclose “exculpatory” evidence, the law remains dangerously unsettled on whether this right applies before a defendant pleads guilty. Worse still, the Supreme Court ruled definitively in Ruiz that “impeachment” evidence – information that could undermine a witness’ credibility or reveal misconduct (also known as Giglio material) – does not need to be disclosed before a plea. This creates a massive loophole. It allows a prosecutor to “bluff” by hiding the fact that their star witness is a paid informant or has a history of perjury, thereby presenting a case that appears to be significantly stronger than it actually is. Unaware of these critical weaknesses in the prosecution’s case, the defendant is then coerced into accepting a plea deal to avoid a trial they mistakenly believe they cannot win.

The consequences are predictable. In Ferrara v. United States, 384 F. Supp. 2d 384 (D. Mass. 2005), the U.S. District Court for the District of Massachusetts vacated a plea after discovering suppressed evidence of witness coercion and perjury by government agents, but such post-­plea challenges are rare because pleas often include appeal waivers. Empirical data reveals the scale of the problem. The NRE reports that official misconduct is a contributing factor in 61% of all exonerations, as of 2024. In a system where 97% of cases are resolved by plea, these violations lead to wrongful convictions of defendants, often indigent and relying on overworked public defenders, who plead guilty to avoid decades in prison, only to discover later that hidden evidence could have proven their innocence.

Why does misconduct persist? One reason is structural. Absolute immunity shields individual prosecutors from civil suits for any conduct “intimately associated with the judicial phase” of a case. Imbler v. Pachtman, 424 U.S. 409 (1976). Courts have overwhelmingly extended this protection to cover plea negotiations, making it nearly impossible to hold a prosecutor personally liable for deliberately withholding exculpatory evidence. This problem is compounded by the Supreme Court’s ruling in Connick v. Thompson, 563 U.S. 51 (2011), which makes it equally difficult to hold the prosecutor’s office (the municipality) liable for failing to train its staff, even in the face of repeated Brady violations. With both individual and systemic civil liability effectively off the table, disciplinary actions rare, and criminal prosecutions of prosecutors even rarer, a culture of impunity is created and perpetuated. This allows misconduct – such as overcharging to leverage pleas, delegating Brady reviews to non-­lawyers, or ignoring exculpatory evidence – to go unpunished.

The problem extends beyond prosecutors. Police misconduct also evades scrutiny in a plea-­dominated system. Without trials, coerced confessions (a factor in 14% of all exonerations, according to the NRE) and fabricated evidence remain buried. Internal police records documenting officer misconduct – often kept on secret Brady or Giglio lists – are frequently withheld by prosecutors despite being clear impeachment material. This allows officers with documented histories of lying to serve as the key witnesses in a case. The defendant, unaware that the officer’s credibility could be destroyed at trial, is confronted with what appears to be an ironclad police report and is pressured into pleading guilty.

Racial disparities compound these injustices. Black defendants, already less likely to receive favorable plea offers, face higher rates of prosecutorial misconduct. A 2022 NRE report found that official misconduct was a contributing factor in 70% of exonerations of Black defendants convicted of homicide, compared to 60% for White defendants in similar cases.

The absence of trials creates additional systemic failures. Without cases going to trial, courts cannot set precedents against flawed practices such as bad forensics, improper witness inducements, and unconstitutional searches, allowing these problems to persist across thousands of cases. Defendants who later discover they were victims of misconduct have limited recourse. Post-­conviction relief is often barred by plea validity standards, leaving them trapped by bargains made under false pretenses.

Cognitive biases further enable these violations. Prosecutors, convinced of a defendant’s guilt, may overlook exculpatory evidence due to “tunnel vision” or confirmation bias. In a trial system, adversarial testing would expose these blind spots. In a plea-­dominated system, they go unchecked, contributing to the thousands of wrongful convictions eventually overturned but only after years or decades of incarceration.

The plea-­dominated system not only conceals misconduct but actively perpetuates it. When prosecutors can secure convictions without accountability, when evidence is never tested, when appeals are waived, the incentive to follow constitutional rules evaporates. The result is wrongful imprisonments, shattered lives, and a criminal justice system that has abandoned its foundational commitment to truth.

Beyond hiding individual instances of misconduct, the plea-­dominated system disrupts the entire system of checks and balances designed to constrain government power. When defendants do not proceed from arraignment to pretrial hearings, police misconduct in building the case goes unchecked. When trials occur in fewer than 3% of cases, neither prosecutors nor defense counsel have incentive to investigate facts thoroughly or vet witnesses carefully. Judges, facing overwhelming caseloads, lack motivation to slow the assembly line for a deeper search for truth. Official misconduct and errors, which pretrial hearings often expose, remain hidden in the case files within prosecutors’ offices where plea deals are struck. And because the process grinds forward without juries, the community loses its oversight role, which is the very function the Framers considered essential to preventing tyranny.

Pretrial Detention as Coercion

The injustices of plea bargaining are compounded by another coercive force: the brutal conditions of pretrial detention. The harsh realities of jail serve as a powerful weapon in prosecutors’ arsenal, pushing defendants, guilty or innocent, to waive their right to trial simply to escape immediate suffering. Though legally presumed innocent, detainees endure overcrowding, violence, health risks, and degrading sanitation that make pleading guilty a rational choice for survival, even if it means accepting punishment for crimes they did not commit. This dynamic erodes constitutional rights while perpetuating racial and socioeconomic disparities because marginalized groups are disproportionately subjected to these punitive conditions.

Overcrowding is a primary driver of this coercion, turning jails into chaotic, disease-­ridden spaces that incentivize quick resolutions. More than 400,000 people are held pretrial in the U.S. on any given day, with local jails often underfunded and ill-­equipped for extended stays. The impact on plea rates is dramatic. A 2020 study by Nick Petersen analyzed over 15,000 felony cases across large urban counties, finding that pretrial detainees plead guilty 2.86 times faster than those released. The study attributed this acceleration to the “pains of detention,” i.e., inadequate health care, high rates of infectious diseases, and “dead time” (jail periods that may not count toward a sentence). As one detainee explained, “Doing dead time forces you to plead; if I had plead guilty earlier, I wouldn’t be in here now.”

For low-­income defendants unable to afford bail, these conditions create impossible choices. They face job loss, housing instability, and family disruption while detained. The prospect of trial, with its risk of even longer sentences, becomes unbearable when weighed against immediate release through a guilty plea.

Violence and abuse transform pretrial detention from mere confinement into a regime of fear. Facilities like Rikers Island in New York City exemplify this dysfunction. A 2014 U.S. Attorney’s Office report found a “pattern and practice of excessive force and violence,” including beatings by guards and unchecked gang activity that violated constitutional standards. Vulnerable populations, particularly those with mental illnesses or disabilities, who are overrepresented in pretrial detention, face heightened risks. The desperation this creates is not theoretical. Detainees facing violence have been documented sacrificing their right to trial simply to secure transfer to different facilities, even if it means pleading guilty to crimes they did not commit.

Health risks add another layer of coercion. Overcrowding and poor sanitation in pretrial facilities facilitate outbreaks of infectious diseases. Detainees with chronic conditions often receive inadequate medical care, creating desperation to escape through any means – including false guilty pleas. A 2017 Human Rights Watch report on California jails documented how pretrial detention punishes the poor through violence and health hazards, with six counties spending $37.5 million over two years holding uncharged defendants in these conditions. For detainees watching their health deteriorate while awaiting trial, the prospect of immediate release through a plea, even for a crime they did not commit, can seem like the only rational choice.

Poor sanitation and inadequate nutrition compound these pressures. Detainees in facilities across the U.S. face lice infestations, diarrheal diseases, and malnutrition. Conditions that would be considered cruel and unusual punishment after conviction are inflicted on those presumed innocent. The Prison Policy Initiative’s research confirms that pretrial detention increases conviction probability primarily through guilty pleas, with no net reduction in crime but heightened risks of rearrest due to the destabilization detention causes.

These pretrial conditions represent a systemic failure that mocks constitutional ideals. Defendants legally presumed innocent are subjected to conditions so degrading that they trade liberty for relief from torment, pleading guilty not because they committed crimes but because survival demands it. This is not justice. It is coercion masquerading as choice.

The Principal-­Agent Dilemma in Defense Representation

Pretrial detention coerces pleas through physical suffering. But even defendants who secure pre-trial release face another form of coercion – one built into the structure of legal representation itself.

In the American criminal justice system, defendants and their attorneys face a fundamental conflict of incentives. Economists call this the “principal-­agent problem”: the “principal” (the defendant) delegates authority to an “agent” (the defense attorney) to act on their behalf, but the agent’s incentives do not perfectly align with the principal’s goals. Defendants seek to minimize punishment or secure acquittals, prioritizing their liberty above all. Attorneys, however, face competing pressures: professional relationships, overwhelming workloads, and the practical realities of a system where trials have nearly vanished.

This misalignment is particularly acute for indigent defendants who rely on public defenders. The result is that many defendants are steered toward guilty pleas even when a trial might yield a better outcome, leading to wrongful convictions, excessive sentences, and the erosion of constitutional rights.

The pressure on defense attorneys to favor pleas over trials is immense. Public defenders, who represent approximately 80% of state and federal defendants, often operate under crushing caseloads that make thorough trial preparation nearly impossible. For decades, the most widely-­cited national guidelines by the National Advisory Commission on Criminal Justice Standards and Goals, first established in 1973 and endorsed by the American Bar Association, have warned that attorneys should handle no more than 150 felonies or 400 misdemeanors per year to provide effective counsel. Despite this warning, attorneys frequently handle hundreds of cases simultaneously, with some state-­level studies showing workloads that are double or triple this standard, spending as little as 20 minutes per case before advising defendants on whether to accept a plea offer. This reality is even more stark when measured against modern research; a landmark 2023 National Public Defense Workload Study by the RAND Corporation found that even the 1973 standards are dangerously high, concluding that for a constitutionally adequate defense, an attorney’s true capacity is closer to 59 low-­level felonies per year.

This resource scarcity creates a powerful incentive to resolve cases quickly through pleas. Trials demand extensive investigation, witness preparation, and courtroom time that overworked defenders simply cannot provide. As David S. Abrams explains in his empirical analysis of over 42,000 cases from Cook County, Illinois, “overworked and underpaid defense attorneys may prefer the brevity of plea bargains to the odyssey of trial.” This preference aligns with the system’s efficiency demands but often comes at the expense of the defendant’s best interests, pressuring clients to waive their right to trial without fully exploring defenses or challenging evidence.

Professional dynamics compound these pressures. Defense counsel, especially in smaller jurisdictions where the same attorneys, prosecutors, and judges work together daily, must maintain collegial relationships to function effectively. Going to trial strains these relationships, forcing familiar adversaries into confrontational proceedings that can create lasting animosity. As Abrams notes, attorneys’ “peers are not the defendants, but the prosecutors and judges, with whom they often maintain long-­term relationships – providing a potential incentive to avoid creating animosity by going to trial.” For public defenders, who are salaried government employees, there is no financial upside to protracted litigation; their compensation remains fixed regardless of outcome. This arrangement encourages attorneys to prioritize system harmony and caseload clearance over aggressive advocacy, subtly steering defendants toward pleas that may not reflect the actual strength of their case and their best interests.

The consequences of this misalignment are severe. Indigent defendants, already disadvantaged by limited resources and often held in pretrial detention, are disproportionately affected. They accept pleas, not because the evidence against them is strong, but because their attorneys lack the time and resources to mount an effective defense and because the uncertainty of trial, amplified by their attorney’s constraints, seems too risky.

This pressure contributes to the near-­extinction of trials. The NACDL’s 2018 report on the trial penalty reveals that fewer than 3% of federal cases now go to jury trial. Without trials, government overreach goes unchecked: racial bias in charging decisions, suppression of exculpatory evidence, and police and prosecutorial misconduct remain hidden because the pretrial hearings that would expose these problems are bypassed in favor of quick plea resolutions.

Tragically, the principal-­agent problem contributes to the conviction of innocents. Overburdened defenders lack the bandwidth to uncover exonerating evidence, leading clients to plead guilty to avoid the trial penalty, i.e., the substantially harsher sentences imposed after conviction at trial. The numbers are stark: NACDL data from 2015 shows that average post-­trial sentences were more than triple those from pleas in most offense categories. For drug trafficking, defendants faced 14.5 years after trial versus 5.2 years for a plea; for fraud, 6 years versus 1.9 years.

Faced with such disparities and represented by attorneys who lack time to investigate, innocent defendants make rational but devastating choices. Of 354 DNA exonerations documented by NACDL, 11% involved guilty pleas to crimes the defendants did not commit. Marcellus Bradford pleaded guilty to aggravated kidnapping to avoid a life sentence, only to be exonerated by DNA evidence after serving 12 years. Michael Marshall pleaded guilty to armed robbery out of fear, serving four years before DNA cleared him. These outcomes stem directly from attorneys’ inability, or systemic disincentive, to push for trials where evidence could be tested adversarially.

Racial and socioeconomic disparities compound these injustices. Black defendants are less likely to receive reduced charges through pleas compared to White defendants, and indigent status increases plea acceptance rates. Strained by impossible caseloads, public defenders lack the time to litigate motions that could reveal racial profiling or police and prosecutorial misconduct, inadvertently perpetuating systemic biases. As the NACDL report concludes, “the pressure defendants face to plead guilty can even cause innocent people to plead guilty,” undermining the presumption of innocence and contributing to mass incarceration. Over 2.2 million Americans are imprisoned today, many after coerced pleas made with inadequate legal representation.

The principal-­agent problem is not an inevitable feature of criminal defense; it is a consequence of policy choices. Crushing public defender caseloads, inadequate funding, and a system that rewards efficiency over justice create misaligned incentives that betray defendants. But these are choices that can be reversed, as Part 3 will explore.

For now, the evidence is clear: when defendants cannot afford private counsel, when their attorneys are overwhelmed by hundreds of cases, and when the system punishes those who dare to go to trial, the right to counsel becomes a mechanism for coercion rather than protection. The constitutional promise that “in all criminal prosecutions, the accused shall enjoy … the Assistance of Counsel for his defense” rings hollow when that assistance is structurally designed to pressure defendants into accepting guilty pleas, regardless of guilt or innocence.

Judicial Resistance: Judges Speak Out Against the Trial Penalty

Defense attorneys face enormous pressures to prioritize pleas over trials. But they are not alone in their frustration with a system that has abandoned its constitutional foundations. Federal judges, who once wielded sentencing discretion to ensure justice, have watched helplessly as mandatory minimums and the trial penalty transform them into administrators of a coercive injustice machine.

Former U.S. District Judge Kevin Sharp for the Middle District of Tennessee resigned after only six years on the bench. After “jumping through hoops to demonstrate he had the judgment and temperament to sit on a federal bench,” Sharp “promptly learned” that Congress “doesn’t want you to exercise that judgment.” Instead, as Sharp told an NACDL panel, Congress “tells judges what to do through mandatory sentencing laws.”

The breaking point came when Sharp was forced to sentence Chris Young to mandatory life in prison for a nonviolent drug offense. Sharp believed the sentence was grossly unjust but had no discretion to impose anything less. Young, Sharp noted, “ran face-­first into the trial penalty. He could have taken a plea,” which would have resulted in a fraction of the life sentence the law now required.

Sharp’s frustration extended beyond mandatory minimums to the coercive nature of plea bargaining itself. In another case, Sharp conducted a plea colloquy where it became clear the accused “didn’t do what the government said she did.” Under Sharp’s probing questions, the prosecutor eventually admitted “they were prosecuting the sister-­in-­law of the actual target.” Sharp refused the plea and dismissed the charges – an act of judicial courage increasingly rare in a system designed to process pleas, not adjudicate guilt.

Prominent former U.S. District Judge John Gleeson for the Eastern District of New York shared Sharp’s concerns about the trial penalty’s constitutional implications. As Gleeson observed, the Sixth Amendment right to a jury trial is “not much of a right if it costs the defendant 3 times or 2 times or even 1.5 times the amount of time in prison for exercising it.” When the government punishes defendants for asserting constitutional rights, those rights cease to function as protections and become traps.

Gleeson also identified a deeper systemic problem: the entire criminal justice system is structured on the assumption that trials will occur. In grand jury proceedings, for example, motions to suppress evidence and Fourth Amendment challenges are prohibited on the rationale that these issues will be litigated later at trial. Similarly, the standard of proof is merely probable cause, not guilt beyond a reasonable doubt; the formal rules of evidence do not apply; there is no duty to present exculpatory evidence to grand juries; there is no prohibition on hearsay; and there is no confrontation of witnesses – all because the system assumes these deficiencies will be corrected and the matters properly tested at a future trial.

As Gleeson explained, “It is a system deliberately set up to permit ill-­advised charges to come through because of the belief that they’ll be exposed and properly litigated at trial. But because trials have dried up, other aspects of the system need to be questioned. We shouldn’t have a system like ours that allows unsupported charges to be leveled in federal court under the assumption that they’ll be fixed during trial … because there are increasingly few trials in our system that guarantee these rights.”

The arbitrariness of the trial penalty was particularly evident to Gleeson when sentencing defendants. He noticed a stark contrast between prosecutors’ sentencing recommendations after trial versus after pleas. When prosecutors argued for 20 to 25 years after a trial conviction, Gleeson would ask what the last plea offer had been. The answer was typically “10 to 12 years.” Gleeson would then ask: “What did you learn during trial that warrants a ten-­year difference between the sentence you would have been fine with if the defendant had pled guilty and the sentence you’re telling me now is the only just sentence to be imposed?”

Gleeson never received a satisfactory answer. “It’s not just the two or three levels that the Guidelines contemplate as a reward for pleading guilty,” he said. “As a general matter, the system dictates that defendants who go to trial need to pay a higher price.” The trial penalty, in other words, has nothing to do with justice and everything to do with punishment for exercising constitutional rights.

Not all judges have accepted the plea system’s dominance. Judge Kimberly Esmond Adams for the Superior Court of Georgia for Fulton County has fought to preserve trials. In her first year on the bench, she presided over 30 jury trials, which is an extraordinary number in an era where most judges handle fewer than five annually. For the past 13 years, she has averaged 18 to 20 criminal trials each year.

Judge Adams’ experience reveals how deeply the plea culture has embedded itself. As she explained at a NACDL panel meeting, “I had to work to change the culture because many attorneys, both prosecutors and defense attorneys, were not interested in trying cases. They simply wanted to run this plea machine.” Acknowledging that presiding over numerous trials is “very demanding,” Adams concluded: “I don’t know another way to do it…. In many ways, [trials are] the only way we’re able to hold judges, prosecutors, and law enforcement accountable.”

Adams’ observation cuts to the heart of the problem: without trials, there is no accountability. The public cannot observe the government’s evidence, judges cannot assess prosecutorial overreach, and appellate courts cannot correct systemic errors. The plea machine operates in the shadows, and injustices and misconduct remain hidden.

The judicial frustrations expressed by Sharp, Gleeson, Adams, and others point to a fundamental crisis of legitimacy. The trial penalty makes a mockery of Congress’ expressed goals for criminal sentencing and calls into question the very purpose of the criminal justice system.

Consider the case of Bordenkircher v. Hayes, discussed in Part 1. The prosecutor offered Hayes five years in prison for check forgery, implicitly declaring that five years was a fair and just punishment that would achieve Congress’ stated sentencing goals: reflecting the seriousness of the offense, promoting respect for the law, providing adequate deterrence, and protecting the public. Hayes rejected the plea, went to trial, was convicted, and received a life sentence.

The logical question is inescapable. If five years was sufficient to meet these goals during plea negotiations, how could a life sentence also be “sufficient but not greater than necessary to meet these goals,” as 18 U.S.C. § 3553 requires? The answer is that it cannot. One of these sentences must be unjust. Either five years was inadequate and the prosecutor was willing to compromise public safety for efficiency, or life in prison is grotesquely excessive and represents pure retaliation for exercising constitutional rights.

This illegitimacy is recognized not just by judges but by former prosecutors as well. Brett Tolman, Executive Director of Right on Crime and a former U.S. Attorney, voiced this concern in a 2024 letter to Congress in support of the Right to Trial Act (H.R. 7349). He wrote, “The heavy reliance on plea bargains has damaged the legitimacy of our criminal justice system and has led to overly long and disparate sentences for identical conduct, with sentences’ lengths being based solely on a case resolving through a trial or plea.” The bipartisan bill, introduced by Rep. Victoria Spartz (R-­IN) and co-­sponsored by members including Rep. Hakeem Jeffries (D-­NY) and Rep. Guy Reschenthaler (R-­PA), seeks to create a “judicial backstop” against the trial penalty. It would give federal judges the discretion, after a pre-­trial conference, to impose a sentence below a mandatory minimum if they find the gap between the plea offer and the potential trial sentence is “so great as to be coercive.” As of late 2024, the bill has been referred to the House Committee on the Judiciary.

When federal judges resign in frustration, when sitting judges acknowledge that trials are necessary for accountability, when former prosecutors admit the system lacks legitimacy, the message is clear. The trial penalty has corrupted the criminal justice system beyond recognition, transforming constitutional rights into liabilities and justice into a transaction. The question is no longer whether reform is needed, but whether America has the will to restore what has been lost.

Quantifying the Trial Penalty: Data and Disparities

Judges like Sharp, Gleeson, and Adams have witnessed the trial penalty’s injustices firsthand. But the scope of the problem extends far beyond individual cases. National data reveals that the trial penalty is not an aberration; it is systematic, pervasive, and staggering in its magnitude.

The trial penalty is ubiquitous. Using data published by the U.S. Sentencing Commission, the NACDL calculated the disparity between sentences imposed after trial and sentences imposed through plea bargains in 2015. The findings are stark: in 31 of 33 offense categories, defendants who exercised their constitutional right to trial received substantially longer prison sentences than those who pleaded guilty.

The disparities are not marginal; they are massive. For drug trafficking, the average sentence after trial was nearly three times longer than after a plea: 14.5 years versus 5.2 years. For fraud, defendants faced six years after trial but only 1.9 years after pleading, more than triple. Those convicted of using, possessing, or trafficking in firearms received 17.6 years after trial compared to 5.8 years for a plea, a difference of nearly 12 years of life.

Some disparities are even more extreme. Tax offenses averaged four times longer after trial (3.2 years versus 0.8 years). Embezzlement and burglary cases showed eight-­fold disparities: defendants who went to trial on embezzlement charges received an average of 4.7 years, while those who pleaded guilty received only 0.6 years – seven months compared to nearly five years. For burglary, the disparity was 12.5 years versus 1.6 years.

The most shocking multiplier appears in food and drug offenses (which typically involve corporate-­level crimes like fraud, misbranding, or shipping adulterated products rather than street-­level dealing), where post-­trial sentences averaged 10 times longer than plea sentences: four years versus 0.4 years. A defendant facing such a choice must weigh four years of incarceration against less than five months, a disparity so extreme it makes a mockery of any claim that plea bargaining is “voluntary.”

Even for violent crimes, where one might expect more consistent sentencing, the trial penalty looms large. Murder defendants who went to trial received an average of 30.7 years compared to 17.5 years for those who pleaded guilty, 13 additional years for exercising the right to a jury trial. For kidnapping, the disparity was 25 years versus 15.9 years. For robbery, 12.3 years versus 6.0 years. For sexual abuse, 19.0 years versus 9.3 years.

These are not mere statistics. Each year represents birthdays missed, children growing up without parents, relationships destroyed, and opportunities lost. A defendant facing the choice between a six-­year plea offer and a 12-­year post-­trial sentence is not making a free choice. They are being coerced by the threat of doubling their time in prison simply for demanding the trial the Constitution promises them.

Yet even these staggering disparities do not capture the full extent of the trial penalty. The Sentencing Commission data has a glaring limitation: it compares plea sentences and trial sentences for the same lead offense, e.g., robbery plea vs. robbery trial. It cannot account for charge bargaining – the prosecutor’s most powerful tool, where they offer to let the defendant plead guilty to a much less serious crime, one that was perhaps never intended to be the final charge.

This practice, which the ABA’s Plea Bargain Task Force notes “is not systematically tracked, and its impact is not fully known,” effectively hides the true trial penalty from statistical analysis. Consider a defendant charged with robbery who is offered a deal to plead guilty to larceny (the federal equivalent of grand larceny) instead. If he accepts, the 2015 data shows the average sentence is 0.7 years, about eight months. If he refuses the deal and is convicted at trial on the original robbery charge, the average sentence is 12.3 years. That is a more than 17-­fold disparity, dwarfing even the eye-­popping multipliers in the official data.

This means the published statistics – already showing sentencing disparities of three-­fold, eight-­fold, and ten-­fold – systematically understate the coercive power prosecutors wield. The true choice is not between a long sentence and a longer one; it is between a few months in jail for a minor offense or potentially decades in prison for a major felony. When a prosecutor can create a 1,700% sentencing differential, no rational person, innocent or guilty, can make a truly “voluntary” decision under such circumstances.

The numbers tell an unambiguous story. These statistics are not abstract; they are the mathematical proof of coercion. They prove that the American criminal justice system, in practice, no longer sentences defendants primarily based on the facts of their case, their criminal history, or their capacity for rehabilitation, as sentencing statutes require. Instead, it sentences them based on a single, procedural choice: whether they accept the deal or dare to exercise their constitutional right to trial. The Sixth Amendment has been perverted from a shield for the accused into the Sword of Damocles dangling precariously above their head. This is not justice. It is extortion dressed in legal language, a system that secures its efficiency by sacrificing its legitimacy.

Human Costs: Extreme Sentences for Exercising Constitutional Rights

The statistics are staggering. Sentences three times, eight times, even 17 times longer for those who exercise their right to trial. But behind every statistic is a human being, a family, a life irrevocably altered by prosecutorial coercion. The following cases illustrate what the trial penalty looks like in practice, not as abstract multipliers, but as decades of human existence traded for the assertion of constitutional rights.

In the late 1980s, Evans Ray, then 30 years old, was convicted of selling small amounts of crack cocaine on two occasions over a three-­month period. While serving his sentence, Ray experienced a devastating loss. In 1993, his younger sister was killed in a traffic accident while driving to visit him. A fellow prisoner and friend, Timothy Patterson, helped Ray through his grief.

Ray was released and rebuilt his life. By 2004, at age 44, he was married and operating his own business with his wife, raising their children in a middle-­class Maryland suburb. He had overcome his substance abuse and left drug dealing behind. Then Timothy Patterson reappeared.

Patterson claimed he was desperately in need of money and pleaded with Ray to obtain drugs for him to sell. Ray refused. But Patterson persisted, and eventually, Ray relented, agreeing to obtain 62 grams of powder cocaine for Patterson. Ray acted merely as a go-­between and made no money from the transaction.

What Ray did not know was that Patterson had recently been arrested for drug dealing and was now working as a confidential informant for the DEA, hoping to secure leniency. When Patterson delivered the powder cocaine to his handlers, they were unsatisfied. The DEA agents insisted Patterson obtain crack cocaine from Ray instead, a distinction that would carry far harsher mandatory minimum sentences.

Patterson returned to Ray and pleaded again. Ray resisted, but his friend’s desperation wore him down. He obtained 60 grams of crack cocaine for Patterson to sell. Again, Ray acted only as a middleman and made no profit.

Ray was arrested. The prosecution offered him a deal. Plead guilty, testify against others, and serve approximately 20 years. If he refused, the prosecutor threatened to seek an enhanced mandatory life sentence based on Ray’s two prior drug convictions from the 1980s, convictions for selling small amounts of crack cocaine.

Ray insisted on his right to trial. The prosecutor made good on his threat, filing the required notice under 21 U.S.C. § 851 to trigger the mandatory life sentence upon conviction. Ray’s first trial ended in a mistrial. At the second trial, he was convicted.

At sentencing, Judge Alexander Williams confronted the starkness of the trial penalty. As the judge observed, “[Ray] of course made a choice to go to trial. No one criticizes anyone who goes to trial, but when he made the decision to go to trial, what the government did, of course, was to initiate the enhancement papers on him and subjected him to life if he were convicted.”

Judge Williams refused to impose the mandatory life sentence, believing it was cruel and unusual punishment. But the government, more concerned with its power than with justice, moved for a corrected sentence. At the resentencing hearing, the prosecutor attempted to deflect responsibility: “This is a mandatory life sentence … because Congress in its wisdom or whatever has decided that a two-­time felon, a two-­time drug felon is subject to a mandatory life sentence. That’s the law, and we can’t change the law.”

Judge Williams was not persuaded, but he was powerless. Forced to comply with the law, he imposed the life sentence while stating his objections for the record: “I believe the bottom line in this case is that a sentence of life under these circumstances would be cruel and unusual – I do believe that – however, I can’t look beyond the [case law] … I am obligated to adhere to the law…. [T]he only reasons the defendant is subject to life is because the government put the enhanced notice on him and based on his record of two previous convictions.”

Evans Ray, who had rebuilt his life and committed no violence, will die in prison for acting as a middleman in a single drug transaction with a government informant, not because the law required it, but because he dared to go to trial.

Evans Ray’s story is not unique. Alice Marie Johnson, a young mother struggling financially, agreed to act as a “telephone mule” in a drug operation, i.e., relaying messages between parties. She never touched drugs, never sold drugs, and had no prior criminal record. Because of Johnson’s minor role and first-­offense status, prosecutors offered a plea bargain: three to five years in prison in exchange for her cooperation and guilty plea.

Johnson, admitting that her involvement was “the biggest mistake of [her] life,” consulted her attorney, who mistakenly believed she could not be convicted of conspiracy since she was never found in physical possession of drugs. On his advice, she rejected the plea offer and exercised her right to trial. In response, prosecutors charged her with offenses carrying mandatory minimum sentences. Johnson was convicted and sentenced to life in prison without parole, plus 25 years.

The disparity was staggering. Three to five years offered, life without parole imposed –more than 10 times the plea offer for a first-­time, nonviolent offender whose role was limited to relaying telephone messages. Johnson would serve 21 years in federal prison before President Trump granted her clemency in 2018, finally releasing her at age 63. By then, her children had grown to adulthood without her, and two decades of her life had been consumed by the trial penalty.

In 1992, Robert Rose found himself in a violent confrontation with his mother’s boyfriend. In the struggle, Rose wrestled a gun away from the man and shot him, killing him. The circumstances suggested possible self-­defense, but the prosecution charged Rose with murder.

While out on bail for three years, on his attorney’s advice, Rose rejected a plea offer of three to nine years’ imprisonment, believing he could prove his case at trial. He was convicted and sentenced to 25 years to life in prison. Rose served approximately 22 years before his release on parole in 2017. Had he accepted the plea, he would have been released 13 to 19 years earlier, years with his family he could never reclaim.

These cases demonstrate the trial penalty in human terms. Evans Ray: 20 years offered, life imposed. Alice Marie Johnson: 3 to 5 years offered, life without parole imposed. Robert Rose: 3 to 9 years offered, 25 to life imposed. If the plea offers represented just and fair punishment – sentences sufficient to achieve Congress’ stated goals of reflecting the seriousness of the offense, promoting respect for the law, and protecting the public – then the sentences imposed after trial are grotesquely excessive, imposed not for justice but as retaliation for daring to exercise their constitutional right to a trial.

And if the plea offers were inadequate – if they compromised public safety for the sake of efficiency – then prosecutors were willing to let dangerous criminals walk free to avoid trial. One of these positions must be true, and either one reveals a system that has forfeited legitimacy. The trial penalty is a dagger in the heart of justice, and Evans Ray, Alice Marie Johnson, and Robert Rose are among its countless victims.

The Ultimate Injustice: Convicting the Innocent

Evans Ray, Alice Marie Johnson, and Robert Rose received grotesquely excessive sentences for exercising their right to trial. But at least they were, in fact, guilty of the crimes charged. The trial penalty inflicts an even more perverse injustice when it coerces innocent people to plead guilty to crimes they did not commit.

Perhaps no single factor destroys the legitimacy of a criminal justice system more than convicting and punishing the innocent. Allowing the guilty to go free is neither fair nor just, but convicting and punishing the blameless is something far worse. It is a betrayal of the fundamental promise that justice makes to its citizens. It turns the Blackstone Ratio – which posits that it is better that 10 guilty individuals escape, than that one innocent suffer – on its head, preferring a system in which it is better that 10 innocent individuals suffer, than that one guilty escape.

The true scale of this injustice remains unknown. Political, legislative, and judicial barriers, rooted primarily in antipathy toward postconviction claims of innocence, prevent determining with precision how many factually innocent people have pleaded guilty. But the known numbers are staggering.

The Innocence Project, which has helped exonerate 375 people using DNA analysis and other scientific methods, reports that 11% had pleaded guilty to heinous crimes they did not commit. The NRE, which has documented case histories since 1989, found that 18 % of the more than 3,700 exonerated individuals had pleaded guilty. In New York alone, 23 of the 321 exonerated defendants had agreed to make false guilty pleas.

Researchers estimate that between 2% and 8% of all convicted felons falsely pleaded guilty to crimes they did not commit. With over 2.2 million people in American prisons, as Senior U.S. District Judge John L. Kane observed, “that is a haunting amount of injustice.” Even using the most conservative estimate (2%), that represents 44,000 innocent people currently imprisoned because they were coerced into false guilty pleas.

The mechanics of how innocent people are coerced into false pleas become clear by examining individual cases. In 2006, George Alvarez was awaiting trial on a public intoxication charge when he was suddenly accused of assaulting a prison guard. Alvarez was innocent – the guards had actually attacked him – but he faced a 10-­year mandatory minimum sentence if convicted. Terrified, Alvarez pleaded guilty. He spent four years in prison until surveillance video, which prosecutors had concealed from him, proved his innocence and secured his exoneration.

James Ochoa faced a similar impossible choice. Charged with carjacking and armed robbery, he was explicitly threatened in open court by a judge with a 25-­years-­to-­life sentence if he dared to go to trial and was convicted. The alternative, a plea for two years, was presented as his only escape. This judicial threat was so coercive that Ochoa pleaded guilty, a decision he made even though pretrial DNA testing on a ski mask had already eliminated him as a contributor. In effect, he was forced to bet his life on a system where the judge’s threat of the trial penalty carried more weight than seemingly dispositive exculpatory forensic evidence. The injustice was later revealed to be even deeper. Unbeknownst to Ochoa at the time of his plea, a deputy district attorney had actually asked the lab analyst to alter the exculpatory DNA report before releasing it to the defense, an order the analyst courageously refused. Ochoa was later exonerated when the DNA was matched to another person who confessed to the crime.

These cases, from George Alvarez and James Ochoa to Rodney Roberts and Keith Brown, reveal a disturbing and repeating pattern. They show the trial penalty is not a neutral, abstract concept but the specific, concrete weapon used to force a plea. It is the judicial threat of “25-­to-­life” that renders exculpatory DNA evidence irrelevant. It is the terror of a 10-­year mandatory minimum that coerces an innocent man to plead guilty to a crime that, as hidden video later proved, never even happened. It is the exploitation of a defendant’s youth, intellectual disability, or indigence to extract a confession and eliminate the risk of a trial. This pattern demonstrates that the trial penalty is not just a tool for efficiency; it is a weapon to secure convictions regardless of guilt, burying the truth and institutionalizing the ultimate injustice.

Pretrial detention adds another layer of coercion that pushes the innocent toward false pleas. Rodney Roberts was a young father with a good-­paying job at a New Jersey mall when he was arrested in 1996 on suspicion of sexual assault and kidnapping. Held in pretrial detention for months, he lost his job and his apartment. When his attorney finally told him about a plea deal, Roberts had 25 minutes to decide.

“I felt like I was choosing between two evils … sabotaging and saving myself at the same time,” Roberts said. Facing a life sentence if convicted at trial, he pleaded guilty. Roberts spent 18 years in custody, both in prison and later in civil confinement, before DNA evidence exonerated him in 2014. The trial penalty, combined with the desperation of pretrial detention, had convinced a factually innocent man that pleading guilty was his only path to eventual freedom.

The trial penalty’s coercive power is especially devastating when combined with false confessions extracted through aggressive police interrogation. Marcellus Bradford was charged with aggravated kidnapping, rape, and murder. Police coerced him into a false confession, and then, prosecutors offered a deal. Plead guilty to kidnapping alone and avoid a potential death sentence. Bradford pleaded guilty and was sentenced to 12 years in prison. DNA testing later confirmed he was never involved in the crime.

In 1989, Tyrone Day faced a similar dilemma in Dallas. Charged with sexual assault, he was confronted with the ultimate trial penalty: a 99-­year sentence if he was convicted by a jury. His own attorney, instead of preparing for trial, provided disastrously mistaken advice – plead guilty, and he would be eligible for parole after just four years. Weighing the certainty of a 99-­year sentence against the hope of release in four, Day accepted the plea. He served 26 years behind bars before DNA evidence finally proved his innocence, and he was exonerated in May 2023.

Michael Marshall’s case demonstrates how prosecutors use the trial penalty to secure a conviction even when their case is weak. Marshall was charged with aggravated assault, armed robbery, possession of a firearm during a felony, and possession by a convicted felon, a stack of charges that guaranteed decades in prison. The prosecutor offered a deal to plead guilty to a single, much lesser charge of theft. As Marshall later wrote to the Georgia Innocence Project, he “plead[ed] guilty out of being scared.” This is a classic example of coercive charge bargaining, i.e., the threat of a decades-­long trial penalty was used to frighten an innocent man into accepting a four-­year prison sentence for a crime he did not commit. Later, DNA testing proved his innocence.

Once police extract a false confession, through threats, exhaustion, or deception, the trial penalty weaponizes that coerced statement into a conviction. Defendants know that juries are unlikely to acquit in the face of a confession, even a false one. The choice becomes: plead guilty and receive a shorter sentence, or go to trial with your own false confession as the prosecution’s star witness and risk decades more in prison.

The trial penalty’s cruelty is most evident when it targets the vulnerable. Davontae Sanford was 14 years old when Detroit police subjected him to coercive interrogation tactics in 2007, questioning him for days without a parent or attorney until he gave a false confession to a quadruple homicide. Facing the trial penalty of life in prison as an adult, Sanford pleaded guilty. He spent nearly nine years in prison. The injustice is staggering because just two weeks after Sanford was sentenced, a professional hitman, Vincent Smothers, confessed in detail to the exact same murders. The state, however, allowed Sanford to languish in prison for years until his exoneration in 2016.

Keith Brown, a 34-­year-­old man with an intellectual disability, was arrested in 1991 for rape and kidnapping in North Carolina. During intense police interrogation, officers fed him details of the crime, producing a false confession. Facing the possibility of a death sentence – the ultimate trial penalty – Brown pleaded guilty in 1993 to avoid execution and was sentenced to 35 years in prison. DNA testing in 1997 excluded him as the perpetrator and identified the actual assailant, leading to his release after nearly four years. North Carolina Governor James B. Hunt Jr. later pardoned Brown based on his factual innocence.

John Dixon’s case shows how multiple systemic failures combine to trap the innocent. He was arrested in 1991 based on a mistaken eyewitness identification, the only evidence against him. The victim had described her attacker as having numerous facial sores, a detail police and prosecutors ignored because Dixon had no such marks. Despite this clear exculpatory fact, Dixon pleaded guilty in July 1991 to multiple charges, fearing a sentence of up to 50 years if convicted by a jury. He later moved to withdraw his plea before sentencing, arguing it was induced by fear of the trial penalty, but the court denied his request, noting that DNA testing was not yet a common practice in New Jersey. Dixon was sentenced to 45 years with a 15-­year parole disqualifier. DNA testing on the rape kit, finally completed in 2001, proved he could not have been the perpetrator. Dixon was released after serving 10 years, a decade stolen from an innocent man by the trial penalty’s coercive force.

The cases of Alvarez, Roberts, Bradford, Day, Marshall, Ochoa, Sanford, Brown, and Dixon are not mere anecdotes. They represent only a fraction of the discovered injustices, the vanishingly rare instances where a lucky break, persistent attorneys, or the power of DNA testing managed to overcome a system designed to bury its mistakes. For every one of these exonerations, how many thousands more remain imprisoned, their innocence forever concealed by the very guilty plea they were coerced into making? That plea, after all, is not just an admission; it is a waiver of the right to appeal, a termination of the discovery process, and a legal seal that forecloses almost any future review.

This is the trial penalty’s ultimate, perverse function. It does not merely punish the exercise of constitutional rights. It transforms the criminal justice system into a machine for manufacturing false convictions. When the innocent plead guilty en masse because the alternative is too terrifying to contemplate, the system has abandoned any claim to legitimacy. Justice is not merely delayed or compromised. It is inverted, punishing the blameless while calling it the administration of justice.

Why the Innocent Plead Guilty: The Innocent Defendant’s Dilemma

The stories of George Alvarez, Rodney Roberts, Marcellus Bradford, and others raise an urgent and troubling question. Why would innocent people confess to crimes they did not commit? Common sense suggests they would not, yet the evidence definitely proves they do, in staggering numbers. Social scientists have studied this phenomenon extensively, and their findings reveal a disturbing truth. Pleading guilty while innocent is not irrational; it is a predictable response to the trial penalty’s coercive calculus.

When faced with the choice between pleading guilty or risking trial, most people assume the innocent would never admit to crimes they did not commit. However, the research and actual cases prove otherwise. Both empirical analysis of real-­world exonerations and controlled experimental studies demonstrate that when faced with the uncertainty of trial and the severity of the trial penalty, many innocent people will accept a plea deal. The question is not whether this happens – but why.

Researchers have identified the mechanism behind this phenomenon: the “innocent defendant’s dilemma.” This term describes a rational-­choice paradox that is a classic problem in game theory, the formal study of strategic decision-­making. In this high-­stakes “game,” the defendant is a “rational player” forced to choose a strategy, while the prosecutor controls the rules and the “payoffs” (the sentences). The dilemma is not a sign of irrationality but a forced calculus of risk versus reward. An innocent defendant, who knows they did nothing wrong, must still weigh the certainty of a lenient plea against the uncertainty of a trial, where a wrongful conviction would result in a catastrophically harsher sentence. As the following analysis shows, this calculus is driven by both brutal mathematics and powerful cognitive biases.

An innocent defendant facing a plea offer of three years in prison must weigh that certainty against the risk of trial. Even if they believe they have only a 10% chance of being wrongly convicted, going to trial carries the risk of a 40-­year sentence, the trial penalty imposed on those who assert their innocence. In game theory terms, the defendant is forced to adopt a “minimax” strategy, one that seeks to minimize their maximum potential loss. A 40-­year loss is catastrophic, while a three-­year loss is merely terrible. Therefore, the rational choice is to plead guilty to “minimize the maximum” damage. The uncertainty of trial, combined with its catastrophic potential outcome, makes the certainty of the plea offer the mathematically “safer” bet, even for the innocent.

Cognitive biases amplify this dilemma. Research in behavioral economics, particularly prospect theory, shows that how a choice is framed dictates the decision. Prosecutors understand this instinctively. They do not frame a plea as a “certain loss” of three years of one’s life; instead, they frame it as a “certain gain” of 37 years saved from the 40-­year trial sentence. Humans are psychologically wired to protect perceived gains, and the effect of this framing is to make defendants lock in the plea rather than “gamble” on trial, even when going to trial might statistically be in their favor.

The psychological pressure intensifies when defendants lack information. This triggers a powerful bias known as “ambiguity aversion” – the cognitive tendency to prefer a known, certain risk over an unknown, ambiguous one. The plea offer is a known (though painful) result. The trial, by contrast, is a complete unknown. In the foregoing abstract example, we assumed the defendant knew their chance of conviction was 10%.

In reality, defendants rarely have such clarity. This debilitating ambiguity is systemic. It is exacerbated by overworked public defenders who lack the time and resources to fully investigate cases. It is also entrenched in the legal system via the Supreme Court’s ruling in Ruiz, which ruled that prosecutors are not required to disclose evidence that could impeach their witnesses before a plea. This manufactured uncertainty, where a defendant does not know if the prosecution’s case is strong or built on lies, makes the certainty of the plea (even if unjust) feel like the only safe harbor. The defendant is not just betting on their innocence but on their lawyer’s ability to prove it, a lawyer they may have spoken with for only 20 minutes.

This theoretical dilemma has been confirmed in controlled experiments. In 2013, Professor Lucian Dervan, former Chair of the American Bar Association’s Criminal Justice Section, conducted a foundational study using a “cheating paradigm.” Researchers placed college students in a lab where they were tempted by a confederate to cheat on a logic puzzle. This experimental design allowed the researchers to know which students were factually innocent (they refused the help) and which were guilty (they accepted it). Afterward, an authority figure accused all participants of academic misconduct and offered them a choice: accept a “plea offer” of a minor, certain penalty (such as an ethics seminar) or risk a formal hearing, the “trial,” which carried a much harsher, uncertain punishment (such as suspension or failing the course). The results were shocking: 56% of innocent students chose to accept punishment for cheating rather than risk the formal proceeding.

A follow-­up study by Professors Miko M. Wilford, Gary L. Wells, and Annabelle Frazier produced similarly damning results. College students were falsely accused of unauthorized collaboration on a test problem and given a choice: sign an agreement accepting 20 hours of “community service” or face formal charges with punishments ranging from probation to a failing grade. More than 80% of guilty students accepted the deal – but so did 52% of innocent students. This finding demonstrates the plea’s catastrophic failure as a diagnostic tool because a defendant’s decision to plead guilty in this scenario provided almost no information about their actual guilt or innocence.

When researchers asked why innocent students confessed, the answers echoed those of wrongly convicted defendants in the criminal justice system. “It didn’t seem like I had much of a choice” and “I didn’t believe I could prove my innocence.” The trial penalty, even in a low-­stakes academic setting, was powerful enough to induce false admissions from the factually innocent.

More sophisticated research, funded by the National Science Foundation, has confirmed and refined these findings. Professor Miko Wilford’s team used immersive simulation scenarios – virtual environments that moved beyond low-­stakes academic misconduct to have participants face realistic plea decisions involving felony-­length prison sentences – to test which factors drive false guilty pleas. This method allowed researchers to model the “calculus of risk” with stakes that more closely mirror the life-­altering choices defendants actually face.

When the gap between a plea offer (e.g., 15 years) and the potential trial sentence (e.g., 55 years) is large, innocent individuals become “increasingly likely to plead guilty.” This finding is crucial because it demonstrates the trial penalty’s coercive power. It effectively erases the distinction between the guilty and the innocent. Both groups are heavily influenced by the sentencing disparity, not by their actual guilt. The trial penalty transforms the question from “Did I do it?” to “Can I survive the risk of asserting innocence?”

The laboratory findings are sobering enough. But in the real world, the innocent defendant’s dilemma is compounded by an additional coercive force: pretrial detention. The defendant facing these impossible mathematical odds is not making the decision from the comfort of home, aided by family and freedom to meet with counsel. They are making it from a jail cell, legally presumed innocent but treated as guilty.

Pretrial detention functions as a structural weapon. While detained, defendants lose their jobs, their housing, and sometimes custody of their children. They are cut off from support systems and severely hampered in their ability to aid in their own defense by meeting with lawyers, locating witnesses, gathering evidence. The detention conditions themselves – overcrowding, violence, disease – create additional pressure to accept any deal that offers immediate release.

This dynamic creates the “time-­served plea,” which is one of the most common and pernicious forms of coercion. An innocent defendant may have already been incarcerated pretrial for six months. The prosecutor then offers a plea for “time served,” resulting in immediate freedom, but at the cost of a permanent criminal record. The alternative is to remain in jail for many more months awaiting trial, continuing to lose everything, while hoping for vindication. For many, this is not a choice. It is an extortionate demand that is nearly impossible to refuse: trade your claim of innocence for your immediate liberty.

Professor Shi Yan of Arizona State University tested this hypothesis directly in a 2021 study. Participants in a simulation were told they were in jail awaiting trial. One group was informed of a COVID-­19 outbreak in the facility, a proxy for the fear and suffering of real-­world detention conditions. The other group received no such information. The results were stark. In the control group, 44% of “innocent” subjects pleaded guilty to secure release, already a disturbing number. But in the group told of the COVID outbreak, that number jumped to 56%. The external pressure of dangerous, degrading detention conditions operated as an independent coercive force, compelling innocent people to plead guilty at significantly higher rates. This finding provides direct experimental proof that the choice to plead guilty is often not a legal calculation but one based on the calculus of survival, made simply to escape the immediate, life-­threatening dangers of confinement.

The laboratory experiments demonstrate what can happen. They prove, in a controlled setting, that the innocent defendant’s dilemma is a powerful and predictable psychological trap. When the sentencing gap is large enough, or the conditions of confinement are sufficiently miserable, the foregoing studies show that a significant percentage of innocent people will rationally choose to plead guilty to minimize their maximum loss.

Data from real-­world wrongful convictions confirms this theory in devastating practice. It proves that this does happen, on a massive scale, and reveals the specific, potent mechanisms of coercion that the labs can only simulate, chief among them the two-­stage trap of a false confession and the overwhelming threat of the death penalty.

A 2023 study published in Psychology, Crime & Law analyzed nearly 3,300 cases from the NRE, finding that 25% were the result of false guilty pleas. The study’s most critical finding was that false guilty plea cases were four times more likely to involve a false confession than cases where innocent defendants were wrongfully convicted at trial.

This reveals a devastating two-­stage process of coercion. First, coercive police interrogation tactics extract a false confession. Second, that confession, which is nearly impossible to suppress and devastating in front of a jury, becomes the ultimate trial penalty. Knowing that juries rarely acquit defendants who have confessed, defense attorneys advise their clients (often correctly) that their chances of acquittal are near zero. The false confession creates a psychological “point of no return,” where the plea is no longer the “safer” option but the only option to avoid a devastating sentence.

Analysis of Innocence Project DNA exonerations reinforces this finding, identifying another specific coercive mechanism. Innocent exonerees who had pleaded guilty were more likely to have been threatened with the death penalty than those who were wrongfully convicted at trial. This provides direct empirical evidence that the most extreme form of the trial penalty – the threat of execution – is highly effective at inducing false guilty pleas, forcing a defendant to choose between a certain life sentence (via a plea) and a possible death sentence (via trial).

The evidence – from controlled experiments, simulation studies, and analysis of thousands of real exonerations – is overwhelming. Pleading guilty while innocent is not an irrational anomaly; it is actually a predictable, rational response to impossible choices. The coercive mechanisms vary: the mathematical paradox of sentencing differentials, the psychological manipulation of framing effects, the two-­stage trap of false confessions, the overwhelming threat of death sentences, and the daily torment of pretrial detention. But the pattern remains consistent. When the risks of asserting innocence become too great, even those who know they are innocent will confess to crimes they never committed.

The implications are staggering. The American criminal justice system is not securing guilty pleas because defendants are guilty. It is securing them because defendants are terrified, because they are rational actors making impossible calculations, because the trial penalty has made the truth irrelevant. This is not the administration of justice.

Part 3: Restoring the Right to Trial

Part 2 documented the trial penalty’s devastating consequences: wrongful convictions of the innocent, racial disparities in sentencing, the erosion of constitutional rights, and a system that prioritizes coercion over justice. The question now is whether these injustices can be remedied and whether America has the political will to act.

There is reason for cautious optimism. Growing dissatisfaction among judges, defense attorneys, reform prosecutors, and advocacy organizations has sparked a movement to eliminate coercive plea bargaining and restore the right to trial. This part examines the reform efforts already underway, from courageous judicial decisions to legislative proposals, and the comprehensive recommendations from legal scholars and advocacy groups to fundamentally restructure how America administers criminal justice.

Judicial Resistance

Judicial pushback is beginning. In March 2025, Senior U.S. District Judge Jed Rakoff issued a notable ruling declaring a portion of the U.S. Sentencing Guidelines unconstitutional. The decision in United States v. Tavberidze, 769 F. Supp. 3d 264 (S.D.N.Y. 2025), strikes at the heart of the trial penalty’s mechanics.

At issue was Guidelines § 3E1.1(b), colloquially known as “super acceptance of responsibility.” This provision offers defendants a one-­point sentencing reduction, typically shaving months or years off a prison term, if they plead guilty quickly enough that the government can “avoid preparing for trial” and “allocate resources efficiently.” The reduction is only available if prosecutors file a motion requesting it, giving prosecutors sole discretion over who receives the benefit.

Judge Rakoff ruled that this system violates the Sixth Amendment in two ways. First, by penalizing defendants who exercise their right to trial with longer sentences, it creates an unconstitutional trial penalty. Second, by allowing prosecutors rather than judges to determine eligibility for the reduction, it unconstitutionally transfers judicial sentencing power to the executive branch.

As Rakoff observed, requiring defendants to plead guilty “promptly,” i.e., quickly enough to save prosecutorial resources, has nothing to do with demonstrating genuine acceptance of responsibility for a crime. Worse, as the NACDL pointed out in its analysis, rushed plea bargains “tend to insulate the government from the consequences of failing to identify and disclose exculpatory evidence.” The pressure to plead quickly prevents defendants from discovering that prosecutors may be hiding evidence of innocence.

Judge Rakoff is not alone in pushing back against the trial penalty. Judge John Kane of the U.S. District Court for the District of Colorado has earned a reputation over decades for scrutinizing plea bargains that appear coercive. In April 2025, Kane made a stunning announcement in open court that signaled his intention to ignore the trial penalty entirely.

In a case involving two defendants with negotiated plea agreements, Kane told the defendants and their counsel that the upcoming sentencing proceedings would not be a “rubber stamp” of the plea deal. The sentence he imposed would be “completely independent of any plea bargain recommendation.” Most dramatically, Kane declared he would “not use the [G]uidelines at all” but would instead sentence the defendants “consistent with the general principles dictated in the United States Code,” the statutory purposes of sentencing that require punishment to be “sufficient but not greater than necessary” to achieve justice. United States v. O’Brien, No. 23-­cr-­00100-­JLK (D. Colo. Apr. 17, 2025); United States v. Ready, No. 23-­cr-­00366-­JLK (D. Colo. Apr. 17, 2025).

In effect, Kane announced he would disregard both the Guidelines’ calculated ranges and prosecutors’ recommendations, exercising the independent judicial discretion that Congress intended but that the trial penalty has eroded. By sentencing based solely on what justice requires rather than what the plea negotiation produced, Kane directly challenged the prosecutorial power that has come to dominate the system.

Kane’s 2025 announcement was not out of character. He has challenged coercive plea practices for years. In 2012, Kane refused to accept a plea agreement that included an appellate waiver, a provision, increasingly common in plea deals, requiring defendants to forfeit their right to appeal. As Kane wrote in his unpublished order rejecting the plea agreement: “Indiscriminate acceptance of appellate waivers undermines the ability of appellate courts to ensure the constitutional validity of convictions and to maintain consistency and reasonableness in sentencing decisions.” United States v. Vanderwerff, No. 1:12-­CR-­00069-­JLK-­1 (D. Colo. June 28, 2012).

That 2012 order, like the 2025 announcement, reflected Kane’s fundamental belief that judges must exercise independent judgment rather than merely ratify deals negotiated in prosecutors’ offices. Together, Rakoff and Kane represent a judicial resistance movement of federal judges using the discretion they still possess to chip away at the trial penalty’s foundations.

But judicial action alone cannot solve a problem created by mandatory minimums, prosecutorial charging decisions, and systemic resource imbalances. District court decisions like Rakoff’s and Kane’s, while symbolically important, are not binding precedent for other courts. The government will almost certainly appeal Rakoff’s ruling, and even if it survives, it applies only within the Southern District of New York unless higher courts adopt it nationally. More comprehensive reforms – legislative, systemic, and cultural – are essential to restore the constitutional right to trial.

Second-­Look Resentencing: Correcting Past Injustices

While upfront reforms like greater transparency in plea negotiations or legislative limits on sentencing differentials are crucial to curbing future injustices, they do nothing for the thousands already serving extreme sentences imposed under the trial penalty’s regime. These are defendants who rejected plea offers years or decades ago and now languish in prison under outdated laws. For them, a different kind of reform is needed: a “backend” mechanism, the opportunity for a second look. This approach recognizes that people can change, that sentences may no longer serve public safety, and that perpetual punishment without review perpetuates the trial penalty’s gravest injustices.

“Second-­look” resentencing is a backend mechanism that allows judges or prosecutors to review and potentially reduce lengthy prison terms after a set period, typically 10 to 15 years. The concept rests on a simple premise: people change. A defendant who rejected a plea offer at age 23 and received a 40-­year sentence is not the same person at age 43. Sentences imposed decades ago may no longer serve public safety or justice, yet without review mechanisms, they remain in force, continuing grave injustices like lifelong incarceration for nonviolent offenses, racial disparities in sentencing, and the denial of rehabilitation opportunities.

Second-­look reforms directly address the trial penalty’s legacy. Defendants who rejected plea offers often faced sentences triple what prosecutors offered, as documented throughout this article. Many of those defendants now serve decades in prison under laws that have since been reformed, but those reforms came too late. Second-­look resentencing provides a pathway to correct these injustices based on current circumstances rather than locking defendants into punishments imposed in a different era.

Second-­look resentencing operates through several mechanisms. The American Law Institute’s Model Penal Code recommends judicial review after 15 years for adults and 10 years for youth offenses, allowing courts to consider rehabilitation, diminished risk, and evolving societal standards. The concept recognizes that extreme sentences fueled by mandatory minimums and tough-­on-­crime policies have led to mass incarceration without enhancing deterrence or public safety.

In practice, second-­look takes different forms across jurisdictions. California’s 2018 expansion of its resentencing law allows district attorneys to initiate reviews for defendants who have demonstrated rehabilitation. Washington, D.C.’s Second Look Amendment Act, enacted in 2020, permits defendants to petition for review after 15 years if they committed their crimes before age 25, acknowledging that young offenders have greater capacity for change. Federally, the First Step Act expanded compassionate release motions, though these are limited to medical or extraordinary circumstances. What is needed is broader second-­look authority that allows review based on rehabilitation and changed circumstances, not just terminal illness or old age.

The scale of the problem is staggering. Over 200,000 people now serve life sentences in the U.S., a population so large it is equivalent to more than 60% of the entire U.S. prison population in 1970. Average federal prison terms have more than doubled since 1988, from 17.5 months to 37.9 months.

Without review mechanisms, individuals remain imprisoned regardless of rehabilitation or diminished public safety risk. Maryland provides powerful evidence. When 188 people who had served 30+ years were released after their convictions were overturned for constitutional errors, only 2.7% were reincarcerated within six years. This demonstrates that extremely long sentences waste resources on low-­risk individuals. Elderly prisoners, who now constitute nearly 20% of the federal prison population, pose minimal public safety threat yet cost an average of $34,704 per year to incarcerate.

The trial penalty makes this worse. Defendants who rejected plea offers received sentences triple what prosecutors offered. Many are now trapped in these excessive terms even as laws have evolved. Congress’ stated goals for sentencing – just punishment, rehabilitation, deterrence, and public protection – are mocked by a system that warehouses reformed individuals simply because no mechanism exists to reconsider outdated sentences.

Criminological research confirms that criminal careers typically last about 10 years. Recidivism rates for released elderly offenders are remarkably low: just 4% for violent offenders released between ages 45-­54, dropping to 1% for those released at 55 or older. Yet without second-­look, reformed individuals remain imprisoned decades beyond any reasonable public safety justification, at enormous cost to families, communities, and taxpayers.

Real-­world implementation demonstrates second-­look’s potential. California’s prosecutor-­initiated resentencing has freed individuals serving life sentences for nonviolent drug crimes who have demonstrated genuine rehabilitation. These releases have reduced prison overcrowding while maintaining public safety because released individuals show low recidivism rates.

Washington, D.C.’s Second Look Amendment Act has made 29% of the district’s imprisoned population eligible for review. Warren Allen, who served over two decades before early release under the law, now contributes to criminal justice reform advocacy without reoffending, exemplifying how second-­look can return rehabilitated individuals to society as productive citizens rather than warehousing them indefinitely.

Federally, the First Step Act expanded compassionate release motions, resulting in a 19-­fold increase in grants – from minimal numbers between 2013-­2017 to 3,691 grants by September 2021. But these releases are largely limited to medical or extraordinary circumstances. Broader second-­look authority is needed to address cases where defendants have been rehabilitated but don’t meet narrow compassionate release criteria.

For second-­look resentencing to fulfill its promise, reforms must include several key elements:

Eligibility: Second-­look should be broadly available without categorical offense exclusions. Excluding violent offenses, as some proposals do, ignores evidence that even individuals convicted of violent crimes show dramatic rehabilitation over time. A 10-­15-­year minimum period served balances finality with the opportunity for review.

Presumptions: Courts should presume in favor of release for defendants who have demonstrated rehabilitation, pose minimal public safety risk, and whose sentences no longer serve congressional sentencing goals. The burden should be on the state to show why continued incarceration is necessary, not on defendants to prove they deserve freedom.

Multiple Pathways: As FAMM advocates, jurisdictions should provide multiple second-­look mechanisms – judicial review, prosecutor-­initiated review, parole board consideration – to prevent backlogs and ensure cases receive timely review.

Procedural Protections: Defendants must have the right to counsel, to present evidence of rehabilitation, and to appeal denials. Victims should receive notification and opportunity to be heard, supporting healing without granting veto power over release decisions.

Data and Accountability: Jurisdictions must collect data on second-­look petitions, grants, and denials, broken down by race and offense type, to identify and address racial disparities.

Federally, Congress should enact legislation allowing defendants to petition for sentence review after 10 years, with courts empowered to reduce terms below mandatory minimums if justice demands. States should expand prosecutor-­initiated review mechanisms and ensure adequate funding for counsel and investigations.

Second-­look resentencing represents a fundamental acknowledgment that the criminal justice system makes mistakes, not just in convicting the innocent, but in imposing sentences that made sense under the harsh regime of the 1980s and 1990s but serve no legitimate purpose today. The trial penalty coerced many defendants into rejecting plea offers and receiving sentences measured in decades. Those individuals have served time, often reformed themselves, and pose minimal risk. Continuing to warehouse them serves neither justice nor public safety.

Implemented thoughtfully, second-­look resentencing restores legitimacy to a system that has abandoned it. Rather than treating sentences as irrevocable regardless of changed circumstances, the system can align punishment with rehabilitation, evolving standards of justice, and the reality that people change. This is not leniency. It is justice delayed but not denied.

Conclusion

This article has documented the trial penalty’s origins, devastating consequences, and potential remedies. But the entire system rests on an inescapable logical paradox. If a prosecutor offers a defendant five years in prison, they have declared that five years is a just punishment satisfying Congress’ mandated sentencing goals. If that defendant then exercises their constitutional right to trial and is convicted, that same prosecutor will often demand a 30-­year, 50-­year, or life sentence. Both cannot be just. One is a lie. Either the five-­year offer was inadequate, meaning prosecutors routinely compromise public safety for convenience, or the post-­trial sentence is grotesquely excessive, imposed purely as retaliation for asserting constitutional rights. This contradiction – that a sentence’s length is determined not by the offense but by whether a defendant dares to demand a trial – exposes a system that has sacrificed justice for efficiency.

This system of coercion was not an accident. It was deliberately constructed. The Supreme Court blessed it in the 1970s, abandoning decades of precedent. Congress armed it with mandatory minimums and sentencing guidelines that transferred power from judges to prosecutors. Prosecutors weaponized it, using threats of life sentences, among other questionable tactics, to extract guilty pleas.

The result is a system where trials, the Constitution’s primary check on government power, are nearly extinct. The grand jury, the motion to suppress, the discovery process, the cross-­examination of accusers – all the constitutional safeguards designed to expose misconduct and test evidence – have been rendered moot. Justice now occurs in prosecutors’ offices through coerced transactions, not in courtrooms through public trials.

The most perverse consequence is not just that the guilty are coerced but that the innocent are convicted. As decades of empirical research and real-­world exonerations prove, the “innocent defendant’s dilemma” is a terrifyingly rational calculation. When George Alvarez faced a 10-­year mandatory minimum for a crime guards committed against him, he pleaded guilty despite his innocence. When Rodney Roberts was given 25 minutes to decide between a guilty plea and life in prison, he traded his innocence for eventual freedom. When faced with pretrial detention’s brutal conditions and the catastrophic risk of decades-­long trial penalties, innocent people will plead guilty to crimes they did not commit. That is not speculation; that is an indisputable fact.

This is the system’s ultimate failure: it manufactures its own victims while leaving actual perpetrators free. A system that convicts the innocent is not broken; it is inverted. It has become the opposite of what it claims to be.

But this system is not irreversible. A broad, cross-­ideological consensus, from the CATO Institute and Right on Crime to the American Bar Association and the NAACP, now agrees that the trial penalty must end. The path to restoring the Sixth Amendment is clear but certainly not easy.

Reform requires transparency: data collection that exposes racial and socioeconomic disparities and coercive practices to public scrutiny. It requires dismantling prosecutorial weapons – repealing mandatory minimums that fuel plea-­bargaining leverage, mandating open-­file discovery so defendants are not forced to plead in the dark. It requires backward-­looking justice, “second-­look” resentencing mechanisms to correct past injustices inflicted on those who dared to go to trial. And it requires judicial courage, empowering judges like Jed Rakoff and John Kane to serve as genuine checks on prosecutorial power, not rubber stamps for pre-­negotiated deals.

These reforms will not come easily. Prosecutors will resist limits on their power. Politicians will fear being labeled “soft on crime.” But the alternative – continuing to coerce pleas, convict innocents, and call it justice – is unconscionable.

The trial penalty is untethered from any coherent principle of justice. It punishes citizens for exercising constitutional rights. It forces the innocent to confess. It perpetuates racial and socioeconomic disparities. It makes a mockery of rehabilitation as well as the notice of justice itself.

Ray Cromartie learned this in the most final way possible. He died on a gurney in 2019 because he believed in his constitutional right to a trial, a right that cost him 22 years on death row and, ultimately, his life. He maintained his innocence. He requested DNA testing that might have proven it. Every request was denied, not because the testing would be unreliable, but because he had rejected a plea offer years earlier. The state killed him not for his crime but for his constitutional choice.

In that same year, the Supreme Court seemed to recall what the system had forgotten. In United States v. Haymond, 588 U.S. 636 (2019), Justice Gorsuch wrote for the plurality that jury trials are “inconvenient for the government.” They are expensive. They are slow. “Yet like much else in our Constitution,” the Court declared, “the jury system isn’t designed to promote efficiency but to protect liberty.”

The trial penalty has operated unchecked for half a century. Its victims – the innocent who pleaded guilty, the guilty serving unconscionable sentences, and the families destroyed by both – are legion. Ray Cromartie died for exercising his right to trial. Rodney Roberts and George Alvarez surrendered their innocence to survive the trial penalty. Thousands more make that same impossible calculation every day. Americans must demand the restoration of a meaningful Sixth Amendment right to trial or continue tolerating a system that punishes citizens for exercising their constitutional rights. The Sixth Amendment guarantees a right to trial. The trial penalty ensures virtually no one dares exercise it.  

Sources: Abrams, David S. (2013). “Putting the Trial Penalty on Trial”; American Bar Association. (2023). “Plea Bargain Task Force Report”; Bibas, Stephanos. (2004). “In the Shadow of the Bench: Judicial Discretion to Reject Plea Agreements.” California Law Review; Bureau of Justice Assistance. (2011). “Plea and Charge Bargaining: Research Summary”; Covey, Russell D. (2007). “The Trial Lottery.” Emory Law Journal; Dervan, Lucian E. (2020). “The Insidious Injustice of the Trial Penalty.” CATO Institute; Dervan, Lucian E. (2021). “Plea Bargaining Law: The Impact of Innocence, Trial Penalty, and Conviction Probability on Plea Outcomes.” Journal of Criminal Law and Criminology; End the Trial Penalty Coalition. (n.d.). “Policy Overview”; Families Against Mandatory Minimums; Hessick, Carissa Byrne. (2024). “Judge Rakoff Takes Aim at the ‘Trial Penalty’.” The Hill; Human Rights Watch. (2017). “Not in it for Justice: How California’s Pretrial Detention and Bail System Unfairly Punishes Poor People”; Innocence Project; Kitai-­Sangero, Rinat. (2017). “A Distant Mirror: American-­Style Plea Bargaining Through the Eyes of a Foreign Tribunal.” William & Mary Journal of Race, Gender, and Social Justice; National Advisory Commission on Criminal Justice Standards and Goals. (1973). “The Defense”; National Association of Criminal Defense Lawyers. (n.d.). “Legislating Freedom: The Importance of ‘Second Look’ Sentencing”; National Association of Criminal Defense Lawyers; Neily, Clark. (2017). “Restoring the Jury Trial.” CATO Handbook for Policymakers; Neily, Clark. (2020). “Organizing a National Movement to End the Trial Penalty.” CATO Institute; Neily, Clark. (2020). “Why the Trial Penalty Must Go.” CATO Institute; Neily, Clark. (2021). “Orwellian Justice: The Trial Penalty Under Fire.” CATO Institute; New York Justice Task Force. (2020). “Recommendations on Second Look Sentencing Reform”; Open Society Justice Initiative. (2011). “Pretrial Detention and Health: Unintended Consequences, Deadly Results”; Petersen, Nick. (2020). “Do Detainees Plead Guilty Faster? A Survival Analysis of Pretrial Detention and the Timing of Guilty Pleas.” Criminal Justice Policy Review; Prison Policy Initiative; Rakoff, Jed S. (2021). “Coercive Plea Bargaining Has Poisoned the Criminal Justice System.” The New York Review of Books; Rakoff, Jed S., et al. (2017). “Trials and Tribulations: The Trial Tax and the Process of Punishment.” The American College of Trial Lawyers; RAND Corporation. (2023). “National Public Defense Workload Study”; Smith, P., Graham, A., Cullen, F. T., Pickett, J. T., & Jonson, C. L. (2025). “Public support for universal second look sentencing, a research note.” Criminology; Spartz, Victoria. (2022). “Spartz Leads the ‘Right to Trial Act’.” Office of Rep. Victoria Spartz; The Sentencing Project. (n.d.). “A Second Look at Injustice”; The Sentencing Project; Tolman, Brett L. (2024). “Letter to U.S. Congress in Support of Right to Trial Act from Right on Crime.” Right on Crime; Ulbricht, Thomas L. V. (2021). “Trials Have Become an Endangered Species.” American Bar Association; U.S. Attorney’s Office, Southern District of New York. (2014). “Manhattan U.S. Attorney Finds Pattern And Practice Of Excessive Force And Violence At NYC Jails On Rikers Island”; U.S. Sentencing Commission; Vera Institute. 

Subscribe today

Already a subscriber? Login