Progressive prosecutors have enacted policies that are soft on crime. They have the chutzpah to claim that their approach is backed up by “data and science,” and that their policies actually reduce crime. They cite a handful of studies, which they hope a gullible public will accept hook, line, and sinker, but those studies are highly questionable, to say the least. The real proof of the unscientific nature of their studies and newly developed data is that crime, especially violent crime, has risen dramatically in their jurisdictions. This hard fact is slowly dawning on the residents in those areas.

At a time when Americans’ trust in scientists has declined substantially since the COVID-19 global pandemic,REF it is peculiar that the George Soros–funded or –inspired “progressive prosecutor” movement confidently asserts that its imaginative reforms are buttressed by science and data.REF The radicals who concocted the movement could not have predicted the COVID-19 global pandemic and the concomitant 14-point decline in trust of scientists compared to pre-pandemic levels of trust.REF Although these prosecutors were not prescient about the fate of scientists’ reputations in the public’s eye, they were aware of a different reality: Millions of Americans are scientifically illiterate.REF

According to a survey by the National Science Foundation, one in four (25 percent) of the 2,200 survey participants did not know that the earth revolved around the sun.REF Armed with this knowledge, advocates of the “progressive prosecutor” (hereinafter “rogue prosecutor”) movement and the district attorneys (DAs) who refer to themselves as such capitalize on the public’s ignorance of science by producing and/or relying on fraudulent “scientific” studies—which are non-replicable and have not been peer-reviewed—which they utilize to buttress their far-fetched claims that their policies are supported by “science” or “science and data” or are “data driven.” The public will not know, much less care, to study their science and data, and the movement simply cites these studies, knowing full well that a compliant left-leaning media will amplify their “results.”

The perversion of the word “science” by the rogue prosecutor movement has indirectly resulted in the unnecessary deaths of thousands of victims of crimes—and has created millions of other crime victims. But for the failed social experiment that is at the heart of the rogue prosecutor movement, this would not have happened in the first place.REF

Science comes in many forms. There is pure science, which is any “system of knowledge that is concerned with the physical world and its phenomena and that entails unbiased observations and systemic experimentation.”REF There is social science, which is the “study of how humans behave and interact within societies.”REF And there is “junk science,”REF which refers “to an argument presented as having been scientifically verified but is false or misleading.”REF

This Legal Memorandum will proceed in two parts. First, it explains the type of data collected by traditional prosecutors across the country, and how and why they use the data. It will demonstrate how the data collected by rogue prosecutors not only differs from data collected by traditional prosecutors, but how rogue prosecutors weaponize incomplete (and sometimes incorrect) data to reduce prosecution rates rather than reducing crime rates. Second, it explains and demonstrates how the “science” that rogue prosecutors use to support their claims is comprised of unreliable, disproven, and/or non-replicable studies. In other words, the data do not support their claims (and occasionally actually refutes them), and their methods for deriving that data are not scientific at all.

A Shaky Foundation

The George Soros–funded or –inspired “progressive prosecutor” movement has been, and continues to be, the most radical, deadly, failed criminal justice social experiment in American history, as chronicled in a recent book written by this author and Zack Smith, Rogue Prosecutors: How Radical Soros Lawyers Are Destroying America’s Communities.REF Despite the fact that several high-profile “progressive” or rogue prosecutors have either been recalled from office,REF removed from office,REF voted out of office,REF or resigned in disgrace,REF the movement continues to be exceptionally well-funded and continues to espouse that progressive prosecutorial policies, such as refusing to prosecute certain crimes or drastically lowering bail, are backed by “data and science” or are “data driven.” The phrases “data and science” and “data driven” are used by those in the movement so often that it seems to be accepted as a given that there is some “there” there; that there are actual data and scientific studies or literature that prove, via accepted scientific methodology, that their these prosecutors’ (non-)prosecution methods reduce crime, support victims, and make communities safer. But it is just not true.

Those inconvenient facts, which they undoubtedly know, have not stopped rogue prosecutors themselves, advocates of the movement, or the media from repeating, ad nauseum, the “data and science” incantation. But stating a false claim over and over does not make it true. Defending flawed methodologies over and over does not make them “data driven.” And citing disproven studies over and over does not make them “scientific.”

Prosecutors’ offices across the country have been collecting data for decades. Such data help them track caseload distribution, the number of misdemeanors and felonies handled by the office, the types of charges filed or dismissed, and other facts and figures relative to keeping the community safe and complying with local regulations and data-retention requirements.

Rogue prosecutors design their data collection efforts to facilitate their pro-criminal, racially discriminatory policies across the office—regardless of the impact on public safety. While they claim that the “science” supports their theory that not prosecuting certain crimes and not asking for bail lowers crime rates, increases public safety, and has not resulted in increased recidivism, a close inspection of those “studies” shows that they are flawed, unscientific, and unreliable. In some cases, they prove the exact opposite—that public safety suffers and recidivism increases when such policies are adopted.REF

How the Data and Science Hoax Is Deployed

The rogue prosecutor movement is animated, as written elsewhere, by the belief that the entire criminal justice system across America is systemically racist.REF It is important to note that there is no one criminal justice “system” in this country: There are 50 states; 3,143 counties; around 18,000 police departments; and approximately 2,300 elected district attorneys. Those facts do not seem to matter to the progressive prosecutor movement. To them, every state system of criminal justice, from New Hampshire to the deep South, to liberal California, to Wyoming and Idaho (the two states with the fewest blacks), is infected with racism.

To “cure” the racism inherent in the system, movement adherents believe, America must “choose prosecutors who will open the locks”REF of prisons and “reverse engineer and dismantle the criminal justice infrastructure.”REF Since their movement is an outgrowth of an earlier movement, inspired by Angela Davis,REF to abolish all prisons, they recruit, fund, and train prosecutors who, once elected, enact sweeping pro-criminal, anti-victim, anti-police policies.

This well-funded, sophisticated movement is prudent enough not to run on their ultimate goal of abolishing all prisons, as they are smart enough to know that the voting public, even uber-liberal denizens of the inner city, would not go that far.REF Instead, they use fuzzy, feel-good sayings like “progressive” and “reform-minded,” and call for Americans to “reimagine” a new type of criminal justice system that relies on restorative justice. They decry so-called mass incarceration, structural racism, the carceral state, the school-to-prison pipeline, and unnecessary incarceration.

The linchpin to this slick sales job is that their newfangled approach is, they proclaim, backed up by data and science. Data and science, taken together, sound important; their cousin, “data-driven,” also sounds important and serious.

The elites in the movement, such as Fair and Justice Prosecution (FJP), realized early that some—or many—Americans are scientifically illiterate and poor at math. As a result, they realized that they could get away with using the science-and-math–sounding phrase “data and science” to justify their approach, and that voters would just nod and trust that there was actual data or real science to back up the expression. With that understanding, it is time to examine how key elected rogue prosecutors deploy the data and science hoax.

• In March 2019, then–Suffolk County (Boston) DA Rachael Rollins argued that 15 different “drug charges, driving offenses, and property crimes” should not be prosecuted; rather, they should be “addressed through diversion or declined for prosecution entirely.”REF Why? Rollins claimed that “recent data” proved her non-prosecution policy was “neither radical nor untested.”REF
• In March 2021, Los Angeles County DA George Gascón announced that, rather than prosecuting and convicting offenders who were “homeless, suffer from substance use or a mental disorder,” he would create a “pre-filing diversion program” to dismiss their cases. In a press release, Gascón claimed he chose these diversion programs, not prosecution, “because the science and data tell us so.”REF Through this so-called science, he stated, “We can truly enhance public safety, increase equity, expand victim services and strengthen police accountability.”REF
• In January 2022, Manhattan DA Alvin Bragg released a memo detailing his priorities as newly elected prosecutor. He argued that there should be “a presumption of pre-trial non-incarceration” with few exceptions.REF Why would a prosecutor argue that the vast majority of offenders should be released from jail before their trials? Bragg claimed that “[d]ata, and my personal experiences, show that reserving incarceration for matters involving significant harm will make us safer.”REF
• In October 2023, Chicago DA Kim Foxx, the first George Soros–funded rogue prosecutor in the United States, issued a press release defending an Illinois law that eliminated cash bail. Foxx argued that “data” supported her decision to throw out cash bail and let more offenders back onto Chicago streets. She claimed that “[i]n this new era of justice reform, our objective remains clear: to ensure a system where detention is determined by risk assessments and not by one’s wallet.”REF

Rollins, Gascón, Bragg, and Foxx are just a few of the rogue prosecutors who claim that “data and science” support their novel approach to criminal justice reform.REF

Proper Data Collection in DA Offices

Last year, FJP, a liberal nonprofit that educates, enables, and supports the broader progressive prosecutor movement, released a report that argued, “An elected prosecutor’s understanding of their office’s work is only as good as the data they collect and analyze. Yet for too long, prosecutor’s offices have been ill-equipped to collect and make use of data, measuring only the most basic things or simply not tracking much at all.”REF

Taking criminal justice reformers at their word, it would be easy to believe that DA’s offices did not collect reliable data until recent memory—but that is not true. Prosecutors have collected and used detailed data for decades.

Since 1990, the U.S. Department of Justice’s National Survey of Prosecutors has required state prosecutors’ offices to submit data about their “resources, policies, and practices.” This required data includes “detailed information on the number of felony cases closed, felony jury trial verdicts, and the use of DNA evidence.”REF As expected, of the 2,300 DA offices across the country (spread across 3,143 counties in 50 states), there is no uniform data collection program or rulebook for prosecutors’ offices to use to collect data. Even within a state, DA’s offices from various counties collect information required by their county and state but do so independently from each other.

But DA’s offices have collected data for decades, well before 1990 when the U.S. Justice Department required them to submit the data mentioned above. Contrary to what FJP asserts, the various DA’s offices are neither “ill-equipped to collect or make use of data,” nor did they collect data for the express purpose of helping criminals. They collected data, mainly, to keep their communities safe, a goal seemingly not shared by rogue prosecutors.

For example, a representative from one large DA’s office in California told the author that his office has been collecting data for decades.REF The data collected include the following:

• All misdemeanors filed by prosecutors, by crime and location.
• All felony reports submitted and filed with the office:
  • Cases dismissed;
  • Cases indicted;
  • Conviction rates;
  • Trials completed and results of each trial;
  • Peremptory challenges made in jury selection (required by state law);
  • Number of inmates housed in jail by category (serving a sentence or pending trial);
  • Types of cases across misdemeanors and felonies;
  • People who re-offend while out on bail;
  • People who received alternative sentences, to include diversion, mental health court, veteran court, domestic violence, and drug court;
  • Trial and plea outcomes; and
  • Success rate of prosecution by category of crime to include all felonies and misdemeanors.
• Recidivism rate across all categories of crime.
• Rate of seeking death penalty and circumstances behind each capital referral or rejection of referral.
• Filing, conviction, acquittal rates for:
  • Domestic violence cases,
  • Sex crimes,
  • Murders,
  • Gang cases,
  • Hate crimes, and
  • Narcotics cases.
• The number of cases submitted and filed by:
  • Category,
  • Conviction rates, and
  • Personnel statistics by Deputy District Attorney.
• Personnel statistics to include:
  • The number of employees by demographic category;
  • Sex of employee; hire date; experience level; trials completed; assignments; awards and recognition.

That office collects scores of other data within the broad categories listed above and additional data required under state law.REF These data enable the district attorney to, among other things:

• Keep the community safe by placing prosecutors in sections of the office that need the most resources;
• Provide training to local law enforcement on the latest legal issues surrounding the Fourth, Fifth, and Eighth Amendments;
• Keep local legislators at the county level apprised of budgetary needs;
• Apply for local, state, or federal grants for specialty programs, such as domestic violence, family violence, drug court, veteran courts, peer/teen courts; and
• Align resources with law enforcement task forces.

A mid-sized DA’s office in the Midwest also responded to the author’s request for the type of data they collect. Like the office in California noted above, this office of less than 50 attorneys has collected data for decades, as far back as anyone in the office can remember. The elected District Attorney told the author that the purpose of data collection in his office is “primarily for staffing, resource allocation, and grant purposes.”REF

Some of the key information they track includes:

• The number of felony and misdemeanor cases presented to the office:
  • The number of felonies and misdemeanors filed or rejected, and
  • The disposition of each case.
• The number of family-involved cases including domestic violence, child abuse, child homicide, elder abuse, intimate partner, and other such cases.
  • The disposition of each case.
• The number and type of all traffic cases and their disposition.
• Victim and witness data is collected to include:
  • Types of cases,
  • Automated updates for upcoming court dates,
  • Restitution and collections of fines, and
  • Contact information.
• Specialty Court Information, to include:
  • Drug treatment court; Mental Health Recovery Court; Veteran Court
    • The number of people considered for placement,
    • The number of people accepted in placement,
    • The number of people who actually completed the program, and
    • The number who failed the program and why.

This data is used by the office in various ways, including, but not limited, to justify staffing requests, to help bolster grant applications, and to satisfy certification or recertification of specialty courts. The office also operates several diversion programs, including a juvenile division, which requires detailed and intensive data collection on participation and success rates. And the office also works closely with local and state law enforcement to identify crime trends so they can proactively create strategies aimed at reducing crime in the community.

Finally, in this author’s discussions with two DAs who led smaller offices (20 or fewer prosecutors), each confirmed that he, too, collects data for the express purpose of public safety.REF Both collect data similar to the mid-size office discussed above and use that data to allocate resources, justify budget requests from the state legislature, and provide transparency to the public. Like the mid-sized and large DA offices, these DAs do not collect data to help criminal defendants or any of the other goals of the rogue-prosecution data collection schemes.

Data to Track Crime Trends

Another way prosecutors use data is to track crime trends. Knowing which kinds of crime are happening in one’s county and where those crimes are taking place is information DAs need to know to do the best they can to advance public safety. Determining how crime rates are changing, as well as how quickly they are changing, can help prosecutors focus their attention on combatting specific offenses at specific times.REF

Data about which staff attorneys try which cases help traditional prosecutors efficiently allocate resources. A county DA can examine each attorney’s individual caseload, and different groups of attorneys’ caseloads, to determine who needs more or less work. For example, if a team of prosecutors trying felony cases has a higher-than-average caseload, the chief prosecutor can assign another attorney to that group. As early as 2001, a National Academies’ report noted that “[d]ata on arrests, caseloads, and conviction rates can be developed and used for short- and long-term planning and resource allocation.”REF

None of this should come as a surprise. All successful businesses assess data, allocate resources, and deploy human capital across the business enterprise to maximize efficiency and achieve their mission, whether that mission is to maximize profits for private/public for-profit companies, stretch donors’ dollars for charities, or, in the case of government offices, make the best use of tax dollars to ensure the delivery of essential government services. The government service the elected prosecutor is designed to deliver, and should deliver, is public safety—achieved by holding criminals accountable. Data collection is essential to ensure public safety and accountability.

How Rogue Prosecutors Weaponize Data

While progressive prosecutors collect some of the same data as traditional prosecutors, in part because local laws often require it and/or to justify their budget requests, they apply the information they gather differently. Rogue DAs use data as a weapon against opposing viewpoints, rather than a tool to accomplish their public duties. This progressive weaponization of data takes two main forms.

Preventing Enforcement of the Law: George Gascón. First, rogue DAs leverage findings about line prosecutors’ cases to prevent them from doing their jobs; they use data like a sword of Damocles,REF requiring line prosecutors to recommend shorter sentences, prohibiting them from adding enhancements or allegations, preventing them from seeking more than one charge in any case, compelling them to seek low or no bail in many cases, and more. In some instances, these policies ran contrary to state law. For example, George Gascón was sued by the Association of Deputy District Attorneys for Los Angeles County (ADDA) over four of his special directives that required these attorneys to ignore or violate state law. The ADDA represents over 800 deputy district attorneys in the Los Angeles District Attorney’s Office, the largest district attorney’s office in the United States.REF

Gascón’s four directives ordered deputy district attorneys (DDAs) to withdraw or dismiss all sentence enhancements or sentencing allegations, including those required under the state’s Three Strikes Law, in pending cases and not to file any such enhancements or allegations in future cases. His directives required his DDAs to dismiss pending gang enhancements, firearm allegations, and certain other “felony prior” enhancements. In its civil suit seeking a writ of mandamus and injunctive relief against Gascón, the ADDA argued that these directives violated California law, which imposed a mandatory duty on prosecutors to plead and prove prior “strikes.”REF

Of note, the California District Attorneys Association (CDAA) filed an amicus curiae (“friend of the court”) brief in support of the ADDA, claiming that Gascón’s directives placed “his deputies in an impossible situation. In order to comply with the directives, each deputy must abandon the laws instituted by this state designed to bring measured justice based upon the facts of each case and must further ignore the voices and interests of the victims of crime despite constitutional duties to the contrary.”REF The CDAA argued that Gascón’s directives attempted to “erase California statutes from application in his county.”REF Echoing a theme repeated throughout the Rogue Prosecutors book,REF the CDAA intoned that prosecutorial discretion “does not translate to an ability to usurp separation of powers lines and permit the District Attorney to undo laws passed by the Legislature and approved by the Governor.”REF

On February 8, 2021, Los Angeles Superior Court Judge James C. Chalfant ruled in favor of the ADDA, issuing an injunction that put a halt to Gascón’s directives that:

• Required DDAs not to plead and prove prior strike under the Three Strikes Law;
• Compelled DDAs to move to dismiss prior strike or any existing sentencing enhancements in pending cases without adequate legal grounds to do so;
• Compelled DDAs to move to dismiss or withdraw special circumstance allegations that would result in a life-without-parole sentence without adequate legal grounds; and
• Compelled DDAs not to use proven special circumstances for sentencing.REF

Judge Chalfant declined to enjoin Gascón from preventing DDAs from charging sentencing enhancements in new cases where not required by the Three Strikes Law. Gascón appealed.

The California Second District Court of Appeals affirmed in part and reversed in part.REF The court agreed that the ADDA would likely prevail on the merits on its claim that Gascón’s directive prohibiting DDAs from pleading prior qualifying convictions violated the state’s Three Strikes Law. Gascón has appealed to the California Supreme Court, and the case is still pending.

George Gascón’s luck ran out on November 5, 2024, when he lost his reelection bid to challenger Nathan Hochman. Hochman ran against Gascón’s policies and lambasted his approach to criminal justice reform. Voters overwhelmingly agreed with Hochman, who defeated Gascon 61 percent to 38 percent on election night.REF

Enforcing “Progressive” Prosecutorial Decisions: Steve Descano. Second, rogue prosecutors use data to make prosecutorial decisions based on race and class, rather than decisions based on public safety. Progressive groups are open about their desire to wield data to control line prosecutors. FJP asserts that “[b]eing able to compare case outcomes across attorneys, including use of diversion and alternatives to incarceration, can also give managers important measures of each attorney’s effectiveness in achieving office goals.”REF On the surface, that sounds reasonable, since all line prosecutors are supposed to support the office’s goals. That is why young law school graduates join prosecutors’ offices—to prosecute cases and keep the community safe. But that is not the goal of rogue prosecutors’ offices.

What do progressive prosecutors do with these measures of “effectiveness?”REF FJP claims that if line prosecutors “are evaluated by how many cases they divert or refer to services, or on other metrics, they will respond to that scrutiny as well.”REF FJP suggests using “multiple metrics…to incentivize line staff to consider varied—and sometimes competing—priorities.”REF Practically speaking, what does that mean? Progressive prosecutors want to use data to change how line prosecutors behave. They can reward assistant DAs for implementing progressive agendas, and they can punish (and in many instances have punished) them for dissenting.

One case study from FJP discusses how Commonwealth’s Attorney Steve Descano of Fairfax County, Virginia, forced assistant DAs to fall into line.REF Descano wanted to release the majority of non-violent criminals facing prosecution in Fairfax without bond.REF But the evidence suggests his line prosecutors disagreed. As FJP notes, “data showed that line prosecutors were recommending detention for nonviolent offenses more often than DA Descano believed was necessary to protect public safety.”REF

What did Descano do? He demanded obedience, ordering “systematic re-training efforts” to rein in his office’s line prosecutors and bring them closer to his policies.REF He used data to corral, command, and constrain assistant prosecutors. And since all line prosecutors in the Fairfax County Commonwealth’s Attorney’s Office are at-will employees, those line prosecutors knew that if they did not toe the line, they could (and most likely would) be fired.

And it is not just Descano’s office in Fairfax pressuring line prosecutors to fall in line with progressive prosecutorial policies. Miriam Krinksy, the Executive Director of FJP, argued that actively measuring how aggressively the line prosecutors implement a rogue DA’s policies “is a key step toward tracking how prosecutors are fulfilling a new vision of justice.”

FJP says the quiet part out loud: It tells prosecutors to “adopt performance standards that reflect your values…. [P]rosecutors should encourage desired outcomes by adopting metrics like reducing incarceration, pretrial detention, and recidivism…. Include these measures in promotion decisions.”REF Put simply, the rogue prosecutor’s playbook is to measure whether line prosecutors are progressive enough, and to use that data to decide who does—and does not—get promoted.

The Cost of Compliance. Until the rogue prosecutor movement came into being in 2017, law students and lawyers joined district attorney’s offices because they wanted to be traditional prosecutors. Being a prosecutor meant enforcing the criminal law, protecting victims, and keeping their community safe by holding criminals accountable. Career prosecutors tend to be rule-followers and by-the-book, commonsense people. So, when a new “progressive” boss is elected and issues new orders that require line prosecutors to ignore the law or, worse, violate the law, or defy common sense in ways that they know will result in increased crime, that compliance becomes an issue.

Line prosecutors know from experience, for example, that not asking for cash bail for violent career felons is a bad idea and will likely result in that felon committing more crimes before trial. They know, from experience, that charging people with resisting arrest deters people from resisting arrest. They know that charging people with possession with intent to distribute drugs dissuades people from dealing drugs. And they know, from experience, that adding gun enhancements to crimes when a criminal uses a gun in the commission of a crime dissuades others from using guns. Yet rogue prosecutors have issued orders eliminating cash bail and prohibited prosecutors from charging people with resisting arrest, from charging people with possession with intent to distribute drugs, from adding gun (or any other) enhancements in a criminal prosecution, and many other “progressive” defendant-friendly policies.REF

Those policies have an immediate and devastating impact not only on the community, but on the traditional law-and-order prosecutors who joined the office. Those line prosecutors have had to make tough choices: Follow the new directives even if they disagree with them or leave the office. George Gascón, however, took things to another level, not just directing line prosecutors to follow bad policies, but directing them to take actions that ran directly contrary to state law. In that case, as described above, several line prosecutors followed a third option, suing their boss seeking a court order permitting them to do their jobs without violating the law.

In order to enforce their edicts (lawful or not), and to send a loud message to anyone who would dare push back on them, rogue prosecutors have used a variety of tactics: firing line prosecutors, muzzling them, forcing them to work in remote satellite offices far from their homes in order to punish them with a hellish commute, and other unprofessional, underhanded tactics.REF Rogue prosecutors also know that by hiring “woke” attorneys who are sympathetic to defendants and are true believers, it will create a work environment in which regular law-and-order prosecutors, who joined the office to fairly but firmly hold criminals accountable, will grow frustrated and just leave. This has happened in many of the 70-plus rogue prosecutors’ offices across the country: The “silent nudge” is effective.

For example, Philadelphia DA Larry Krasner, upon assuming office in 2018, simply fired 31 career prosecutors, many of whom were in the homicide division.REF Prior to being elected, Krasner had been a civil rights attorney, defense attorney, and public defender. On the night he was elected to office, he proclaimed that he was now a “public defender with power.” No doubt, he guessed that the career prosecutors in his office would not like and might not cooperate with his “new vision,” so he directed his deputy to fire them forthwith. In response to a question about why he fired these career prosecutors, a spokesman for Krasner said that he was “given a clear mandate from the voters for transformational change…[and the firings] are necessary to achieve that agenda.”REF According to The Philadelphia Inquirer, in his first year in office, 261 attorneys left the office.REF More shocking is the fact that over 70 attorneys Krasner hired have since left, which, according to the local paper, has resulted in a state of chaos.

Race- and Class-Based Decisions. In addition to using data to intimidate line prosecutors, rogue DAs use office data to make prosecutorial judgments based on race and class, rather than law and justice. Criminal justice reform nonprofits make this point explicitly. FJP argues that “[e]nsuring that racial and ethnic data are properly tracked is obviously only a starting point for prosecutors committed to alleviating systemic racism and the harm caused by the legal system.”REF

It states that “[w]henever possible, research and reports should disaggregate data by race and ethnicity, emphasizing that the costs of the status quo are disproportionately borne by people of color.”REF Once progressive prosecutors have that data, they use it to make prosecutorial decisions. Take, for example, this counterintuitive warning from FJP: They caution prosecutors from considering a defendant’s arrest history when deciding whether to offer them bail, because it “reinforce[s] patterns of racial disparity.”REF

The Brennan Center for Justice, a far-left advocacy organization, is even clearer. In an article titled, “The Data that Can Make Prosecutors Engines of Criminal Justice Reform,” they argue that rogue DAs should use data to redefine prosecutorial “success” along the lines of race and equity. They state, “In trying to redefine what success in prosecution looks like in the 21st century, data is key.”REF And how is data key? The Brennan Center claims that “in prosecution, the metrics of success often end up reflecting the number of convictions, sentence length, or even crime rates. However, these factors do not reflect a community’s health and wellbeing. They do not tell us, for example, about community trust, support for survivors, or if outcomes are fair across racial and ethnic groups.”REF

FJP and the Brennan Center are not trying to hide the ball. They want elected officials to use “the data” to redefine “justice” along racial lines.

And how exactly do you measure community trust, health, or well-being? By broadening the goal of the criminal justice system from enforcing the criminal laws of the state and holding criminals accountable—its narrow but noble purpose—to capacious concepts like solving health disparities, ending homelessness, or increasing community “well-being”—whatever that means. The Brennan Center and others like it, including the rogue prosecutor movement, engage in a dangerous sleight-of-hand by signaling to the community that the criminal justice system is in place to solve all of society’s problems. It is not, and never was designed to do so.

However, that has not stopped them from using the criminal justice system, and, in particular, data collection, to further their social goals. Take, for example, Alan Bragg in Manhattan. In a letter to his staff, Bragg stated that his office would consider the “the racially disparate use of incarceration” when recommending sentencing guidelines.REF Or consider George Gascón in Los Angeles. On the masthead of his campaign website, Gascón pledges to “reform our justice system so it works for everyone.”REF Does he mean that his reforms are designed to keep communities even safer, protect victim’s rights, and hold criminals accountable in new and creative ways? Far from it.

The Los Angeles DA states that, instead of evenly and fairly prosecuting crimes regardless of the race of defendants, he will spearhead “reforms to reduce racial disparities” in Los Angeles prisons.REF Simply put, progressive prosecutors are putting radical policy proposals into practice and are using their office’s data collection schemes to enforce their policies on their subordinates. Far from judging criminal offenders by the “content of their character” or the crimes they allegedly commit, rogue prosecutors make judgments based on the data they collect and the color of citizens’ skin.REF

How Rogue Prosecutors Get the Data Wrong

Not only do rogue DAs misuse data in the attempt to rein in assistant prosecutors and reduce or eliminate racial disparities, but they also often back up bold claims with bad data. Progressive prosecutors’ writings are ripe for critique because they contain concerning amounts of unsupported, inaccurate, or misleading claims.

Consider Manhattan DA Alvin Bragg. In a 2020 memo to his staff, he claimed that “[s]tudies show that even three days in jail can lead to a loss of housing, employment, and strain family connections and increase the likelihood [of] failure to appear in court.”REF Bragg cites a single article to support this claim, but takes a quote from the executive summary of the study completely out of context to bolster his thesis. The quote from the study is this: “Some researchers and legal professionals believe there is a relationship between the number of days spent in pretrial detention and the defendant’s community stability (e.g., employment, finances, residence, family), especially for lower-risk defendants.”REF

Herein lies the problem: The study he cited does not support that proposition. Rather, it focused on the impact of pretrial detention on pretrial outcomes, specifically, the likelihood of the individual committing new crimes while awaiting trial or failing to appear for subsequent court dates and post-disposition recidivism.REF It had nothing to do with the loss of housing, employment, or finances.

A second progressive prosecutor, George Gascón, has made similar baseless claims. After taking office, Gascón wrote that “[s]tudies show that prosecution of the offenses driving the bulk of misdemeanor cases have minimal, or even negative, long-term impacts on public safety.”REF Readers never received a chance to learn which studies show that misdemeanor prosecution is ineffective, because Gascón does not cite any.REF The rogue prosecutor failed to back up his claims.

Organizations composed of progressive prosecutors exhibit a similar failure to cite data upon which they claim to rely. FJP claims that “[s]ome research shows that increasing the number of minority prosecutors in an office decreases racial sentencing disparities.”REF FJP at least provides an endnote, but it simply says that: “[s]teps for diversifying office staff include developing targeted recruitment to diverse groups (like bar association affinity groups); reassessing hiring criteria to address barriers to hiring people of color; and ensuring that underrepresented groups on staff are appropriately supported, considered for promotion, and involved in office hiring decisions.”REF

Regardless of whether those steps are good ideas, the endnote does not cite any studies showing that the proposed steps decrease sentencing disparities. FJP also claims that “[m]ost people in jail in the United States are there because they can’t afford bail” without citing any supporting source,REF and that traditional ideas about prosecutorial success “contributed to the explosion of mass incarceration from the 1970s through the 1990s”—again without citing a source to support this claim.REF These omissions are important. The reasons that people are in jail and the reasons prison populations increased in the late 20th century are points of real dispute between progressive and traditional prosecutors.REF

In sum, progressive prosecutors use data the wrong way. They weaponize it to curb line prosecutors. They use it to reduce racial disparities in the courts, rather than prosecuting crimes based on the evidence.

Rogue Prosecutors and “Science”

Having made clear what rogue prosecutors mean when they use the word “data,” it is now important to consider their use of the word “science.” First, what do rogue prosecutors mean when they say “science”? Second, how do they reference this so-called “science”? And finally, are the studies that they cite actually scientific—or are they disproven by better research? The evidence will show that rogue prosecutors use a limited group of empirical studies to defend their radical policies. A closer analysis of two such studies will reveal an uncomfortable truth: The “science” these prosecutors cite is often incorrect.

When rogue prosecutors claim that they follow “the science,” they do not mean that they are conducting experiments in a lab by using Bunsen burners or test tubes or microscopes. Rather, rogue prosecutors reference studies that compile empirical data. These studies ask questions such as, “If judges release criminal defendants before their trials without bail, do they commit more crimes?” Or “If prosecutors refuse to charge misdemeanor offenses, how does it impact their counties?” Empirical studies gather data over extended periods of time to provide concrete answers for these difficult questions. And that is a good thing, because questions about bail, recidivism, non-prosecution, and local crime deserve answers. Correct, verifiable answers.

How do data scientists ensure that their studies are correct? At the very least, empirical studies should compile objective evidence and produce replicable results. Science is a method to ensure that experiments and the data derived from them are reproducible and valid.REF The scientific method is a set of procedures and practices, the aim of which is to provide valid data.REF

Two studies that are heavily relied upon by rogue prosecutors are neither factual nor replicable. The first study argued that reducing cash bail in Cook County, Illinois, did not put Chicago residents in danger. The second study argued that refusing to prosecute misdemeanor offenses in Suffolk County, Massachusetts, did not lead to increased crime rates in Boston.

Case Study One: The Chicago Bail Study Hoax. The Chicago study about reduced bail reveals that some rogue prosecutors are willing to use faulty, disproven research to defend their offices’ practices. In 2017, Judge Timothy Evans, the Chief Judge of Cook County, Illinois, at the behest of Kim Foxx, the Cook County State’s Attorney, issued an order changing bail practices in his circuit and increasing the number of defendants released before trail.REF In 2019, Judge Evans issued a report on the data he collected. The Illinois Circuit Court made a bold claim about Evans’ findings: “The report shows that the increased release of defendants from jail did not increase the threat to public safety in Cook County.”REF Simply put, the court argued that letting more and more criminal defendants back onto Chicago streets before their trials would not put law-abiding citizens in danger.

Evans’s report concluded that there was no significant increase in violent crime after judges began implementing reforms to either reduce or altogether eliminate cash bail for many pretrial defendants.REF The report touted findings that only a “small fraction” of those released were charged with a new violent offense.REF To illustrate the point, the report trumpeted that only 147 felony defendants (0.6 percent of the total) who were released from custody in the 15 months after bail reform went on to be charged with new violent crimes.REF If true, that was truly impressive, and helped make the case that serious bail reform, à la the Chicago way, was a viable, perhaps meritorious, way forward for other cities to follow.

After the study came out, the Chicago Council of Lawyers issued a statement saying that bail reform had “been a tremendous success.”REF

Here’s the problem: Not only was it not a success, but it was a complete and utter failure. First, a Chicago Tribune team exposed myriad holes in the study’s conclusions and published a lengthy expose on Evans’s flawed study.REF Among the many flaws in the study, the paper found:

• Evans’s definition of “violent crime” was limited to six offenses, and excluded others, including domestic battery, assault, assault with a deadly weapon, battery, armed violence, and reckless homicide. The Tribune notes: “Hundreds of these charges were filed against people released after bail reform took place…. If those charges were included in the analysis, the total would be at least four times higher.”REF
• The report’s underlying data was so flawed that it led to an undercount of murders and other violent crimes committed by people out on bail. Evans’s report noted that there were only three such individuals; the Tribune identified 21 people.
• Evans’s analysis only included defendants whose initial charge was a felony and excluded those charged with a misdemeanor, which is far more common. The Tribune found that five of the study’s murder defendants had bonded out of jail on misdemeanor charges—four of whom had past felony convictions from attempted murder to armed robbery—and three had served prison time.
• The Evans analysis only counted the first new charge against defendants after they were released from custody. The Tribune identified two people who were released, charged with another crime, released again, and then charged with murder, all within the time period under question.
• Evans excluded from the study three murder defendants whose first charge occurred before bail reform took place, even though they were released on bond after the reforms took place in September 2017.
• In one instance, a reputed gang member released on a no-cash bond after having been arrested on a felony gun charge allegedly shot and killed a city employee who was driving home to his three children. He was not included in Evans’s analysis because his pretrial release assessment was not entered into the data set.
• At least 2,334 other felony defendants also did not receive a risk assessment and were not part of Evans’s analysis.

When the Tribune presented Judge Evans with its findings, his office stood by its analysis and conclusions, writing, “In no way does this report intend to minimize new criminal activity.”REF But that is exactly what the report tried to do. It was, in fact, the major conclusion—that lowering or eliminating cash bail did not result in a significant increase in new crime, even though, as it turns out, it did.

As if that was not bad enough, the bail study’s methodology was severely criticized and its conclusions were thoroughly debunked by two law professors (one of whom was a federal district court judge) in a Wake Forest University law review article.REF The 45-page article began by criticizing the rudimentary methodology utilized by the study, noting that it “merely” looked at the “change in total number of crimes after a reform and then attributing that change (or stability) to that single factor.”REF The authors noted that “a researcher must consider potential confounding variables that might contribute to any trends.”REF For example, the authors noted that the study’s failure to control for other factors, such as the 2016 Chicago homicide spike, evinced a “serious” methodological problem.

As a way of demonstrating their point, the authors noted that they had published a study that considered changes to the city’s stop-and-frisk policy that occurred in late 2015 and its relationship to the infamous 2016 Chicago homicide spike. Instead of simply looking at before-and-after crime totals, the authors “crafted multiple regression equations controlling for factors that have been reported in the literature so have some association with crimes.”REF The authors noted myriad other factors that should have been taken into consideration by the Chicago Bail Study, including the hiring of police officers in 2017 and 2018; the deployment of federal U.S. Bureau of Alcohol, Tobacco, Firearms and Explosives agents and federal prosecutors who focused on gun crimes; the use in 2017 of new “shot stopper” technology in high-crime neighborhoods; and the newly formed Partnership for Safe and Peaceful Communities, which committed $75 million to reducing gun violence in Chicago.REF

The Chicago Bail Study was so poor that it even failed to control for seasonal changes in crime rates.REF Thus, it is not surprising that the authors concluded that “without any effort to control for these other factors that likely reduced crime in Chicago during the after period but not the before period, it would be unreasonable to assert that pretrial release changes did not increase crime in Chicago.”REF

Turning to the conclusions of the Chicago Bail Study, the authors found that “contrary to the Study’s suggestion of stability, the number of crimes committed by pretrial releasees appears to have significantly increased.”REF The authors found that “the number of released defendants charged with committing new crimes increased by about 45 percent…[and] the number of pretrial releasees charged with new violent crimes increased by about 33 percent.”REF

In other words, the bail lowering experiment not only was a failure, but it also made matters worse. How much worse? The authors noted that the percentage of “aggravated domestic violence prosecutions dropped by prosecutors increased from 56 percent before [bail reform] to 70 percent after.”REF The authors noted that it was reasonable to infer “that the increase in dropped cases resulted from batterers more frequently obtaining pretrial release and intimidating their victims into not pursuing charges at trial.”REF From the author’s experience as a domestic violence prosecutor and homicide prosecutor, it is not only reasonable to make that inference, it would be unreasonable not to draw that conclusion. Properly understood, the actual science shows that letting more people out on bail increases crime. Yet rogue prosecutors continue to peddle this single junk study to defend their no-bail policies.

Cook County’s circuit court later defended its analysis by pointing to a subsequent paper from two Loyola University professors.REF The Loyola report said that Chicago bail reform “had no effect on new criminal activity or crime.”REF What the report meant, it explained, is that the percentage of defendants released pre-trial who committed new crimes remained relatively static both before and after the bail reform measures were put in place (although a slightly higher percentage of defendants who were granted pre-trial release following the bail reform failed to appear for their subsequent court hearing).REF

However, as a later editorial pointed out, that does not prove that the absolute number of crimes stayed the same.REF In fact, the number of overall crimes committed by defendants granted pretrial release must have increased because more defendants were getting out on bail once the bail reform policy was implemented.REF Why? Because if the percentage of pre-trial defendants who committed new crimes stayed static but the number of defendants who got released before trial went up after the bail reform measures were implemented, the absolute number of pre-trial offenses had to have gone up.REF Indeed, the Loyola report bore that out, although the authors concluded that the increases were “within the range that would normally be expected.” Try telling that to the victims of these new crimes and their families.REF

Case Two: The Suffolk County Study. During her tenure as District Attorney for Suffolk County, Massachusetts, Rachael Rollins claimed that a study by Rutgers, Texas A&M, and New York University professors backed up her decision not to prosecute nonviolent misdemeanors.REF The study contends that a “marginal nonprosecuted misdemeanor defendant” is 53 percent less likely to receive “a new criminal complaint within two years post-arraignment” than “‘complier’ defendants who are prosecuted.”REF Put simply, refusing to prosecute misdemeanor offenders makes them 53 percent less likely to reoffend. But just as the Chicago Bail Study was flawed from the start, the Suffolk study made crucial errors.

First, instead of examining all criminal defendants in Suffolk County or even all misdemeanor offenders, the study only considered first-time misdemeanor offenders. Why is that a problem? The study purported to examine whether crime rates will increase or decrease when misdemeanors go unprosecuted, but as any good prosecutor knows, first-time misdemeanor offenders are the least likely criminal defendants to re-offend.REF The question is not whether the first-time (i.e., people with no criminal record) misdemeanor offenders will re-offend—many will not; the question is whether the whole group of people that Rollins decided to let off the hook through her policy would be likely to re-offend. The study she relied upon to support her policy made no such claim about the propensity of non-first-time offenders to reoffend. Examining only the most favorable subset of the data and—even worse—–ignoring the data most likely to refute one’s hypothesis is not remotely scientific.

Second, the Suffolk County study categorized data in a way that produced misleading results. When a prosecutor is determining whether to charge a defendant for a misdemeanor, he can make one of several decisions. First, he can charge. This begins the typical criminal process. Defendants can enter a guilty plea, a no contest plea, or take the case to trial. Second, a prosecutor can drop the charges, exercising the prosecutorial discretion not to bring a case based on the facts and circumstances of that case and that defendant. Historically speaking, prosecutors generally dropped charges because there is not enough evidence to obtain a guilty verdict, but rogue prosecutors sometimes drop charges for other reasons, like reducing racial disparities.

A third, final, and unrelated decision that prosecutors can make is to send a defendant to a diversion program such as rehabilitation for a drug addiction. In a diversion program, defendants admit guilt and agree to participate in the program and its requirements. Defendants who enter diversion programs are still “prosecuted”—the prosecutor allows them an alternative to incarceration, but he can still reinstate charges if the defendant fails to comply with the terms of the diversion program.REF On the other hand, if the person successfully completes the diversion program, that person’s record (with respect to that charge) is expunged or the charges are reduced or dismissed. So while diversion is an alternative to a trial, fines, or prison, there are still consequences. Defendants do not get to go on their merry way.

The Suffolk County study failed to distinguish between cases that were charged, cases that were diverted, and cases that were dropped. Instead, it simply drew a line between “prosecution” and “non-prosecution.”REF How did they draw these overly broad lines? They argued that if a case did “not proceed past the day of arraignment,” or did “not result in a conviction or an ‘admission to sufficient facts,’” then it was never prosecuted.REF

In the first footnote of the study, the authors claim that alternatives to prosecution include “declining to prosecute, diversion, dismissal, pretrial probation, and deferred adjudication.”REF At no point in the study do they distinguish between these radically different outcomes or ask whether diversion programs led to fewer defendants re-offending. The study fails to draw distinct lines between diversion and refusing to prosecute. It masks a radical strategy (non-prosecution or prosecutorial nullification) with a proven strategy (diversion). The Suffolk County study is radically—and perhaps intentionally—vague.

Third, and worst of all, the Suffolk County study data is inaccessible and irreplicable. The Heritage Foundation asked the authors of this study for their data in order to examine it. The researchers responded that they were under non-disclosure agreements (NDAs) with the Suffolk County prosecutors who gave them their data.REF The NDAs prevented them, they explained, from releasing the data so that Heritage could examine it.REF The DA’s office eventually released the researchers from their NDAs: The researchers, however, have not yet released the data in a format such that it can be analyzed.REF

There is currently no way for Heritage—or presumably anyone else—to test the results of this study. The Suffolk study is the quintessential definition of non-replicable, as evidenced by the fact that researchers will not provide the underlying data, and there have been no other reputable studies that have reached a similar result.

Conclusion

Rogue prosecutors defend their radical strategies by leaning heavily on two serious-sounding words, “data and science.” They claim that “the data” supports their decision not to prosecute misdemeanor offenses. They proclaim that “the science” backs their policy preference for no-cash bail and scores of other pro-criminal policies. But these arguments misapply real data and lean on flawed, disproved science. From punishing local prosecutors, to charging based on race, to using false statistics to push dangerous policies, rogue prosecutors abuse the phrase “data and science” to push partisan political ideology.

What makes the phrase “data and science” so compelling is that, in the public’s mind, data and science are objective. They are rigorous. They are reliable. And, by and large, they are determined by experts. Americans would not want to believe that such “objective” measures are being weaponized by prosecutors to force their staff into letting dangerous people back onto the streets or misquoted in policy papers to justify prosecutors’ inaction. But too often they are.

Data and science are not the problem. The problem is the way rogue prosecutors use both words. Saying the words “data” science” over and over may sound credible, perhaps even “great and powerful,” but after pulling back the curtain, one discovers that rogue prosecutors’ appeals to “data and science” are nothing more than a smokescreen—a “data and science” hoax.

Charles D. Stimson is a Deputy Director of and Senior Legal Fellow in the Edwin Meese III Center for Legal and Judicial Studies at The Heritage Foundation. He is the co-author of Rogue Prosecutors: How Radical Soros Lawyers Are Destroying America’s Communities (Bombardier Books, 2023). The author would like to thank John G. Malcolm, Zack Smith, Seth Lucas, Curtis Herbert, and Caleb Sampson for valuable comments and assistance on an earlier version of this Legal Memorandum. Any mistakes are mine.