In a nice post about the Court’s Census decision, Rick Pildes noted that Chief Justice Roberts approached the case in a realist fashion, probing beneath the surface to figure out what actually did and would happen. This realism was entirely absent in the Court’s other end-of-term blockbuster. In Rucho, Roberts was as pure a formalist as can be imagined. He explained that several institutions other than the Court could do something about partisan gerrymandering. But not once did he ask whether it’s likely that they would take action. The answer, of course, is that it’s not likely at all. Gerrymanderers rarely stop gerrymandering voluntarily. That’s exactly why judicial intervention is needed—because the political process can’t be trusted to fix itself.

Consider Roberts’s examples from the states. Yes, it’s true that Florida now has a Fair Districts Amendment barring partisan favoritism in redistricting, and that “voters in Colorado and Michigan approved constitutional amendments creating multimember [redistricting] commissions.” But these reforms were all enacted through voter initiatives, not by self-interested legislators. The legislators actually fought the initiatives tooth and nail, seeking to keep the mapmaking power for themselves. Moreover, while these particular initiatives succeeded, most ballot measures aiming for redistricting reform fail. And they typically fail precisely because of the furious opposition of sitting politicians. In any event, more than half the states have no initiative process (including highly gerrymandered states like Georgia, North Carolina, Texas, and Wisconsin). In these states, there’s no way for people even to vote on anti-gerrymandering proposals.

Or take the “dozens of [congressional] bills [that] have been introduced to limit reliance on political considerations in redistricting.” As Justice Kagan pointed out, in her dissent for the ages, “what all these bills have in common is that they are not laws.” In its entire history, Congress has never directly curbed partisan gerrymandering. Very occasionally, it has regulated redistricting in other ways, such as by requiring the use of single-member districts. But not once has Congress limited the partisan intent or partisan effect of district maps. Nor is it apt to anytime soon, since, in Kagan’s words, “the politicians who benefit from partisan gerrymandering are unlikely to change partisan gerrymandering.”

Roberts’s reasoning in Rucho, then, is plainly formalist rather than realist—focused on what other responses to gerrymandering are possible rather than probable. But his opinion is actually worse than that. Most of the responses he identifies are ones that he himself has previously suggested are constitutionally questionable. His opinion therefore has the feel of Lucy holding the football for Charlie Brown, only to take it away when he tries to kick it.

Consider the state reforms lauded by Roberts: state court decisions, state constitutional provisions limiting the discretion of mapmakers, and independent redistricting commissions. All of them are suspect under Roberts’s own dissenting opinion in Arizona State Legislature v. Arizona Independent Redistricting Commission. That’s because all of them are typically instituted by actors other than the state legislature. In his Arizona State Legislature dissent (which may command five votes now that Justice Kennedy is off the Court), Roberts argued that the Elections Clause empowers the state legislature to design congressional districts—and forbids other state institutions, in particular commissions, from doing so.

Or take the congressional bills cited by Roberts. He’s the author of Shelby County v. Holder, the single decision in modern times most skeptical of Congress’s power to regulate the electoral process. It’s not much of a leap from discounting Congress’s authority to fight racial discrimination in voting under the Fifteenth Amendment to dismissing its right to combat partisan gerrymandering under the Fourteenth Amendment. As if to confirm this suspicion, Roberts pointedly added, “We express no view on any of these pending [bills].” You can bet that when this view is expressed, it won’t be sympathetic.

Roberts’s approach in Rucho, then, is not the true formalism of a justice committed to analyzing institutions based on their official powers. Instead it’s faux formalism: an acknowledgement that other institutions could potentially act, which would be followed by a swift repudiation if they ever did so. “Go ahead and try to fix gerrymandering,” Roberts seems to be telling the states and Congress, aware they probably won’t. “And if you do try, you’ll find the Court blocking you at every turn.”