A little knowledge is a dangerous thing. This maxim comes to mind when journalists who lack any legal background attempt to engage in complicated empirical studies of judicial decisions.

The latest headline from the New York Times is titled "Trump's 'Superstar' Appellate Judges Have Voted 133 to 12 in His Favor."

But the data suggests that in the 13 appellate courts, there is increasingly such a thing as a Trump judge. The president's appointees voted to allow his policies to take effect 133 times and voted against them only 12 times. . . .

The Times analyzed every judicial ruling on Mr. Trump's second-term agenda, from Jan. 20 to Dec. 31 of last year, or more than 500 orders issued across 900 cases. About half of rulings at the appellate level were in Mr. Trump's favor — better than his performance with the district courts, though worse than his record at the Supreme Court, where the rulings on his agenda have almost all been on a preliminary basis in response to emergency applications.

My immediate reaction concerned not the numerator, but the denominator. How many Trump circuit appointees were actually able rule on Trump cases? For starters, the authors do not define what it means to rule "in Mr. Trump's favor." Does that include a random APA challenge to a regulation passed in a prior administration? Or do they count a mundane Title VII case against a federal agency? The authors do not actually share their data set, which makes scrutinizing it impossible. At least academics share their data, which makes it possible to dissemble the studies.

Let's assume the data set is limited to litigation against Trump executive actions. The majority of the anti-Trump litigation has been filed in the First Circuit, where until recently, there were zero Trump appointees. Then there is the D.C. Circuit, where Judges Katsas, Rao, and Walker are the only ones. I can think of a smattering of Fourth and Ninth Circuit opinions where Trump appointees would up on the panels, but that is a small number.

If you read about three-quarters of the way down, you get to what might be called a selection bias in the data set:

Mr. Trump's success on appeal has also been driven by the influence that his appointees have wielded in specific judicial circuits, especially the U.S. Court of Appeals for the D.C. Circuit. The court has jurisdiction over federal matters in the nation's capital, and its three Trump appointees have exercised outsized influence, repeatedly sitting on panels hearing key cases.

Combined, Judges Gregory G. Katsas, Neomi Rao, and Justin R. Walker voted 75 times in favor of the administration — slightly more than half of the pro-Trump votes from Mr. Trump's appointees logged by the Times analysis — and only three times against.

Again, the authors found a total of 133 total votes for Trump, and they attributed 75 to these three judges. Again, I was still perplexed by the denominator. Were these three judges really on that many panels with Trump-related cases?

If you keep reading further, the authors describe their methodology. You learn that the authors count separately a vote for an administrative stay, a stay pending appeal, and the merits:

When Mr. Trump's policies are temporarily blocked by district court judges, appeals courts can issue "administrative stays," temporary rulings that effectively reverse the lower court's orders and let contested policies take effect. Administrative stays are supposed to be temporary but can remain in place for weeks or even months. In many cases, they are replaced by a more lasting stay, known as a "stay pending appeal," that remains in place while the appellate court considers the case.

The Times analysis tracked both kinds of stays, as well as the final rulings that appellate courts made after considering arguments from both sides.

Mr. Trump's nominees sided with him consistently across all three kinds of rulings, voting in his favor 97 percent of the time on administrative stays, 88 percent of the time on stays pending appeal, and 100 percent of the time on final rulings.

So it seems the number of rulings is inflated triply: 75 rulings may break down to 20-something cases. Even on the Supreme Court, a vote to grant interim relief will usually predict the same vote on the merits.

Let's dig a bit deeper. In many of these cases, as I recall, the vote to grant the administrative stay was unanimous. In other cases, the justification to issue a stay pending appeal was made based on Supreme Court precedent. Indeed, Judge Rao dissented in Slaughter, arguing that the majority failed to follow Wilcox and Boyle.  The Times also fails to mention that Judges Katsas and Rao disagreed concerning Judge Boasberg's contempt proceeding. Moreover, how many of these conservative votes were vindicated on appeal--especially by Justices Alito and Thomas, who were not Trump appointees. This limited analysis proves very, very little.

I think what we have here is an small set of emergency docket cases at the D.C. Circuit, where conservative judges, two of whom worked in the executive branch, supported a strong theory of executive power. No surprise there. At the very end, the authors quote Leonard Leo who stated the obvious:

According to this view, Mr. Trump's judicial nominees are doing their jobs by pausing and reversing rulings by district court judges who overreach.

"The Constitution provides for a relatively strong executive," said Leonard Leo, the Federalist Society co-chairman who guided Mr. Trump's first-term judicial picks under the banner of "originalism," which seeks to determine the original public meaning of the Constitution and often generates conservative outcomes. He said it should not come as "any surprise" that Mr. Trump's originalist judges would rule in his favor.

The Times also has the chutzpah to talk about "gamesmanship," and at least create the impression that Rao and Katsas did something underhanded to get on all of the stay panels.

The three judges' prominence in the data is partly a function of the circuit's practice of assigning emergency motions to special three-judge panels. These panels are chosen randomly, on a rotating basis, according to a spokesman for the circuit executive's office. The three Trump appointees were often selected for the panel during the spring and summer, when many judges on the district court it oversees were ruling against the administration.

Judges Katsas and Walker declined to comment; Judge Rao did not respond to a request for comment. Keeping the same group of judges on the panel for weeks, as the court did for much of 2025, can lead to "gamesmanship" by litigants, said Marin K. Levy, a Duke law professor.

I've heard this claim from liberals before. Nonsense. The Chief Judge of the D.C. Circuit is not going to stack conservatives on panels. (The Times could study how often Chief Judge Boasberg was assigned to certain cases.) These stay panels are set well in advance. And if you'd like to learn about judges keeping cases on specific panels, google Boyce Martin and Stephen Reinhardt. For so many writers, the world began in 2017, took a break in 2021, and resumed in 2025.

I would like to see how often district court judges in Boston ruled in favor of Trump on anything of substance. I think the number would be close to zero. The same judges were reversed by SCOTUS multiple times.

I'll close with one final point. The article suggests that all of the Trump appointees are casting their rulings as a form of auditioning.

He has called judges who ruled against his administration "radical" and "lunatic." He has praised judges who rule the way he wants, calling them "highly respected" and "brilliant."

"You could have court of appeals judges auditioning in case a Supreme Court seat opens up," said Morgan Hazelton, a political science professor at Saint Louis University and the co-author of a book on collegiality in the appellate courts.

There are more than 50 circuit appointees. Do you think all of them are auditioning? I would not deny that some might be, but the vast majority of the nominees have no credible shot at promotion. Indeed, a law professor recently wrote on a listserve  that Judge Jerry Smith's dissent in the redistricting case was an attempt to curry favor with Trump. Judge Smith is nearly 80 years old. We need to exit this audition trap. You cannot simply dismiss an argument by saying the judge is auditioning. The truth may be that the judges actually believe what they are writing. For the left, that truth is too hard to process, so they rely on the "auditioning" fiction.

It is not unusual for newspapers to discuss empirical work. Very often, Adam Liptak will discuss empirical analyses from Lee Epstein and other credible scholars. I enjoy Adams's end-of-year surveys. And even when Liptak highlights empirical work, he will quote those who reviewed the study, and take objection. My colleague Jon Adler recently did so with another study profiled in the Times about the Supreme Court apparently favoring the wealthy. The author of this new piece, Mattathias Schwartz, attempted to do his own empirical analysis. And he has done so before. He doesn't share his work with anyone, so no one can figure out what is going on. There is accordingly no quoted criticism of his work, or scrutiny of his methodology. My advice is to take what Schwartz writes with a grain of salt.