An Unsurprising Term from an Unsurprising Court

The Supreme Court is in summer recess following its October 2025 term. I’ve had a chance to reflect on the term, and one takeaway is that the Court decided almost every case as most observers predicted, especially after oral argument. Indeed, Josh Blackman reports that “[i]n the aggregate,” competitors at FantasySCOTUS “predicted 94.74% of the cases accurately [this term].” Yes, sometimes a Justice or two will catch the world off-guard (they are not automatons), and sometimes interesting issues that could divide the Court end up being irrelevant. But overall, each Justice has demonstrated how he or she generally approaches cases, and even when a Justice or two votes in unexpected ways, the Court’s ultimate judgment can often be predicted with pretty good accuracy.  

In response to the rise of AI models that forecast the Court’s decisions (albeit while “systematically overpredict[ing] ideological division”), Justice Sonia Sotomayor laments such predictability, which she says shows the Justices “may not be stepping out of our normal thinking and opening our minds to new ideas enough.” True, there can be value in looking at the law with fresh eyes—one of my favorite things about Justice Clarence Thomas is that he tries to get to the bottom of old, oft-repeated doctrines; as he puts it, “We never go to the front see who’s driving the train, where is it going.” But even such unpredictability can, in a sense, be predictable; we know how Justice Thomas will approach a case.  

Predictability is neither startling nor necessarily problematic—though, obviously, it is better when the Court is consistently correct! It also would be a problem if the Court’s decisions could be predicted based on reasons other than law. My experience with the Court, however, is that the Justices have been thinking about how to decide cases for decades, and they show their work with lots of opinions and votes. Given those conditions, it is unsurprising when patterns emerge. In fact, we should be concerned if there were no patterns. One reason why law is a profession is that lawyers can predict outcomes: “People want to know under what circumstances” they will run into governmental “force.” As the late Justice Antonin Scalia observed:  

Even in simpler times uncertainty has been regarded as incompatible with the Rule of Law. … As laws have become more numerous, and as people have become increasingly ready to punish their adversaries in the courts, we can less and less afford protracted uncertainty regarding what the law may mean. Predictability … is a needful characteristic of any law worthy of the name. 

When methodological commitments are combined with the reality that the Court decides relatively few cases, and that the cases it does decide involve questions that have often been percolating in the lower courts or involve issues similar to those the Court has recently decided, the result is that the Court does not prompt many surprises. Because many of the Court’s cases are difficult and multi-layered, it's impossible to plug each Justice’s voting record into an algorithm and get perfect predictions, but the Court doesn’t throw that many curveballs.     

For example, perhaps the most important case of the term was Trump v. Slaughter, which held that the President can fire agency heads at will, even if Congress says otherwise. Given the nation’s history with that constitutional question (first argued in 1789), scholars may debate Slaughter for centuries to come. But no one can say that the Court’s decision was a surprise. The Court has been telegraphing Slaughter for years.   

The most surprising case of the term, in my opinion, was Watson v. Republican National Committee. There, the Court upheld Mississippi’s law allowing mail ballots to be counted if postmarked by Election Day and received within five business days. Justice Amy Coney Barrett joined the Chief Justice and the Court’s Democratic appointees, who held that federal law requires ballots to be cast by Election Day, not received by Election Day. When the Court granted review, I assumed it would reverse; if you don’t know anything else about a case, the safest bet is to assume the Justices are inclined to reverse. After doing more research, I came around to the other side, and at oral argument, it sounded like a majority would affirm. The Court, however, reversed in a closely divided decision. I’m not the only one surprised by Watson.   

Some of the Court’s major cases were also hard to predict. For example, in Learning Resources v. Trump — the tariffs case — I predicted before oral argument that the Court would conclude that the major questions doctrine does not apply the same way in the field of international relations. That prediction looked shaky after oral argument, and ultimately a plurality disagreed.  

I confess that I also predicted Landor v. Louisiana Department of Corrections and Public Safety incorrectly, at least before oral argument. When the Court grants review without a circuit split, that’s often a signal that the Court wants to correct an error. The case, moreover, was framed as a religious liberty case. Landor is a Rastafarian who, for religious reasons, does not cut his hair, yet when he was transferred to a new prison, officials allegedly threw away a court decision allowing him to keep his hair and shaved him. Bad facts sometimes make bad law, and those are very bad facts for prison officials—especially given the Court’s demonstrated willingness to protect an inmate’s religious liberty. 

Yet, the Court (per Justice Neil Gorsuch) held that “Congress’s power to spend money does not include the power to regulate. Spending Clause statutes can bind only those who voluntarily and knowingly undertake obligations by agreement with the federal government.” Because Congress enacted the Religious Land Use and Institutionalized Persons Act pursuant to the Spending Clause, —as a matter of constitutional law—cannot directlyimpose coercive burdens on private persons. Congress can spend money for the “general welfare,” but imposing personal liability against individual officers (who are not parties to any agreement with the federal government) is not spending federal money. To be sure, the Court identified how Congress—or the States—could allow suits for damages. Doing so, however, would require additional legislation.    

Likewise, although the outcome was previewed during oral argument, I have questions about Trump v. Cook. I am not surprised the Court said that the President cannot direct the Federal Reserve’s exercise of monetary policy. Aditya Bamzai and I have authored an article explaining why history supports such a rule. But I did not expect the Court to resolve the (not squarely presented) constitutional question, and to conclude—apparently—that the Federal Reserve’s current slate of powers is fine: “In upholding the constitutionality of the Federal Reserve as currently structured and with its existing enforcement authorities, we do not suggest that Congress could assign the Federal Reserve additional regulatory powers that are attenuated from monetary policy.” Yet the Federal Reserve today does a lot more than monetary policy, and I don’t understand how the Fed can act as a pure regulator without being subject to the rule from Slaughter that “the President may remove his subordinates at will.”      

There were also some lower-profile surprises. Justice Brett Kavanaugh, for example, is rarely in dissent, yet he dissented in Watson, Learning Resources, and Trump v. Barbara, the birthright citizenship case. And Justice Ketanji Brown Jackson dissented alone in Chiles v. Salazar, the case about whether Colorado can prevent “counseling conversations with minors that might encourage them to change their “sexual orientation or gender identity, including efforts to change behaviors or gender expressions.” The outcome was unsurprising given First Amendment law, but I expected a unanimous decision or a multi-Justice dissent urging a doctrinal change. Instead, there was a solo dissent.    

To be sure, I don’t claim to have followed every case, let alone predicted every outcome. My point, though, is that most cases were decided, more or less, as close observers would expect.  

So, what about next term? Thus far, the Court has only granted 25 cases, many of which also strike me as predictable. For example, in Nielsen v. Watanabe, the Court is likely to overrule an implied cause of action against federal officials. As Justice Barrett just explained, the Court now “reject[s] the practice of fashioning rights of action as we see fit” because “[h]ome-grown causes of action are difficult to reconcile with the Constitution’s separation of legislative and judicial power.” In St. Mary Catholic Parish v. Roy, the Court is likely to hold—applying recent precedent—that Colorado cannot exclude Catholic preschools from a universal-preschool funding program. Given that Justice Kavanaugh has tipped his hand, moreover, the Court presumably will hold that the Second Amendment protects possession of AR-15s. The Court also probably will hold that “parents have standing to challenge a law or policy that deliberately displaces their decisionmaking role as to ‘gender transitions’ of their children” and that parties cannot be held “in civil contempt based on a violation of an injunction’s ‘spirit’ where the injunction is silent as to the conduct upon which contempt is based.” 

The Court is a complex institution run by smart, strong-willed individuals. What direction it takes over time can be surprising, but rarely is that true for individual cases, especially with the benefit of the Justices’ questions at oral argument. Knowing what is likely to happen, however, helps the public plan their affairs. Although that makes opinions less exciting, predictability is also part of the rule of law.  

Aaron L. Nielson is a senior fellow at the Civitas Institute and holds the Charles I. Francis Professorship in Law at the University of Texas at Austin School of Law.