A year and a half in, Trump’s administration still has not filled hundreds of executive appointments. But many observers are equally alarmed by the appointments he has made. New EPA helmsman Andrew Wheeler, Department of Education chief Betsy DeVos, and Department of Interior head Ryan Zinke are pushing ahead with a regulatory agenda that delights movement conservatives but, by prevailing standards of administrative law, looks procedurally flawed, statutorily suspect, and just plain botched.

Writing in The Hill, for example, Bethany A. Davis Noll and Richard Revesz note that the scientific and evidentiary basis legally required for the Trump environmental regulations is largely absent. These legal deficiencies have led some experts to predict that the seeming revolution in regulatory policy will come a cropper, as federal judges on the D.C. Circuit and then the Supreme Court throw out the shoddy work. Courts to the rescue!

The viability of the Trump regulatory revolution is an open question. In two years, the federal judiciary will look quite different than it does today.

But the viability of the Trump regulatory revolution is a more open question than these experts think it is. Their confidence is based on an overly static appraisal of the federal judiciary and administrative law doctrine. Many Trump regulations won’t wind their way into the courts for eighteen months to two years, possibly even longer. At that point, the federal judiciary won’t look the way it does today.

Working closely with the Federalist Society and, in all likelihood, a Republican Senate majority leader, the Trump administration will have a chance—modest, but real—to reshape the D.C. Circuit and, with the nomination of Brett Kavanaugh and perhaps other justices, the Supreme Court. Prevailing case law endorses the so-called Chevron doctrine, which holds that courts should defer to agency decisions on substantive matters if there is a reasonable way to do so, but the reconfigured courts of Trump’s third and fourth years would likely push that doctrine to new and extreme versions. If so, empowered conservatives could uphold almost anything from a Trump agency as lawful. The Court’s recent 5-4 upholding of the Muslim Ban, though based on a different area of law, could foretell what a sufficiently conservative judiciary would do in administrative law.

• • •

How likely is this scenario?

First consider the D.C. Circuit, often called the second most important court in the land. It is the venue for most administrative law cases—the main playing field for judicial review of federal regulations. Its decisions can be appealed to the Supreme Court, and those appeals are sometimes heard when the regulations under review are especially controversial—as in the cases regarding Obamacare. But for less serious policies, the D.C. Circuit’s pronouncements are usually final.

The D.C. bench comprises eleven active judges. Like all circuit courts, it hears cases in panels of three, which are usually chosen more or less randomly from the bench. The panel determines the case disposition—who wins—by majority vote. Extensive studies of circuit voting show a strong partisan effect in politically salient cases (losers in a 2-1 split typically acquiesce silently rather than dissent).

The Democratic majority on the D.C. Circuit—second most important court in the land—is the result of an intense fight by the Obama administration.

Since panels are chosen randomly, the likelihood of a majority-Democrat panel depends on the overall make-up of the bench. Currently seven of the eleven are Democrats. This majority is the result of an intense fight by the Obama administration, which belatedly realized that control of the D.C. Circuit was vital for the fate of the Obamacare and Dodd-Frank regulatory regimes. Republicans—as usual, sharper than Democrats when it comes to judicial politics—had alighted on this fact earlier, systematically blocking Democratic appointments. Their stubborn intransigence forced the Democratic majority to pull the “nuclear” trigger: Senate Democrats revised parliamentary rules that had been on the books for four decades so that a simple majority, rather than a sixty-vote supermajority, can confirm federal judicial nominees and executive appointments. This move allowed President Obama to build a Democratic majority on the D.C. Circuit at the last moment. Now that decision haunts the Democrats.

Thanks to those latter-day Obama appointments, the odds today that a randomly drawn panel from the circuit will have a Democratic majority is about 72 percent. (The details of all calculations in this essay can be found at my website.) That is good news to most liberals, but still about 28 percent of panels drawn from the current bench will go the other way. What if one of them upholds a Trump regulation—or goes even further, using the case to spearhead a Super-Chevron doctrine?

The strong Democratic majority on the bench provides an extra layer of protection against such an outcome. Under circuit rules, any active judge may request a so-called en banc review—a review of the panel’s decision by the full bench. Granting such review requires a majority vote, only active judges may vote, and the outcome of the full bench’s review is again determined by majority rule. Of course, en banc review is very rare. Because it requires a lot of time and effort, and disrupts comity among the judges, it is employed only for exceptionally controversial cases. Still, it happens. It was a majority-Republican panel that struck down critical parts of Obamacare; the bench granted en banc review and reversed the decision.

Under the current composition of the bench, the threat of such reversals will likely keep most Republican-majority panels in line when it comes to assessing Trump policies or articulating new legal doctrine. But for lesser stakes, the Democratic majority will often acquiesce to the 28 percent of Republican panels. The Supreme Court would probably pass on reviewing these moderate- to less-important cases, especially since a Republican-dominated Supreme Court will favor the outcomes. So even today, a considerable portion of Trump policies would likely survive their encounter with judicial review.

• • •

What about two years from now? So far, Trump has filled only one seat on the D.C. Circuit, appointing Judge Gregory Katsas by way of a nuclear-option override of a Democratic filibuster in the Senate. He was confirmed 50 to 48. What are the odds that Trump will be able to replace one or especially two more Democrats on the Circuit?

It is hard to predict retirement probabilities. If instead we apply generic actuarial data to the current Democratic justices, we find that the chance that none dies in the next year is 89 percent; over two years that figure drops to 80 percent. That means there is a 1 in 5 chance that at least one Democratic seat becomes open within the next two years and switches to a Trump appointee. If that should occur, the balance on the bench would become 6 to 5, and the probability of Republican-dominated panels would rise to 43 percent. Trump regulations would be more likely to survive.

If we look six years out, Republican dominance of the D.C. Circuit and the Supreme Court becomes almost certain.

What really matters, though, is the chance that two Democratic judges are lost—since then en banc control would flip to the Republicans. In that case, the Democratic minority would be able to pick off Trump regulations only when it lucked into a Democrat-dominated panel, some 43 percent of the time, and even then the minority ruling could be reversed en banc by the Republican majority.

If we take the actuarial tables at face value, the probability that at least two Democrats expire in a year is about 1.5 percent; in two years, it is a little less than 4 percent. In light of these numbers, liberals might breathe a sigh of relief. True, many Trump regulations would slip through Republican-dominated panels. But the D.C. Circuit looks like a formidable obstacle to sloppily drafted policies from the EPA, the Department of Education, and the Department of the Interior.

Unless the Supreme Court weighs in. Now that Anthony Kennedy has resigned, the current Court has a balance of four liberals and four conservatives. With the likely confirmation of Brett Kavanaugh, this configuration will be a solid 4-5 conservative majority. Given a docket of Trump regulations, will this Court build a Super-Chevron doctrine and use it to overturn the work of the Democratic majority on the D.C. Circuit? It is not impossible, but a radical departure seems unlikely. On the other hand, the situation would be quite different on a 3-6 court with a hard conservative supermajority.

What are the prospects that one of the four Democratic justices will depart over the next two years, opening the door to a radical rewiring of administrative law? The chance that none dies in the next year is 86 percent, according to the actuarial probabilities. Over two years that figure drops to 75 percent. In other words, the chance that at least one does die over the next two years is 25 percent.

Putting the analyses of the two courts together, we see that the prospects aren’t so rosy as has been claimed. If Trump gets his 1 in 4 chance to replace one of the Democratic Supreme Court justices, the game is over: most Trump regulations will almost surely survive, except for the small to moderate ones that Democratic-majority panels on the D.C. Circuit can block without triggering review. And even if all the Democratic justices remain on the Supreme Court, there is a 4 percent chance the D.C. Circuit will be reconfigured so that the Republicans have a strong majority and en banc control. Overall, that’s a 28 percent chance that when it comes to Trump’s regulatory agenda, almost anything will go.

• • •

These back-of-the-envelope calculations provide only rough proxies for retirement probabilities, of course. Most modern justices do retire from office rather than die there. Moreover, small groups of people may deviate strongly from the large-number odds of the actuarial tables. And other factors may prove important. If the Democrats retake the Senate in the 2018 midterms (unlikely but not impossible), the Trump administration can give up hope of reshaping the D.C. Circuit—though who knows what would transpire with a Trump Supreme Court nominee and a Democratic Senate. Garland redivivus?

Trump’s regulatory revolution has a real chance of succeeding even if its craftsmanship makes administrative law experts gag.

Still, the calculations highlight two broad points. First, Trump’s regulatory revolution has a real chance of succeeding even if its craftsmanship makes administrative law experts gag. Its fate hinges on remaking the courts and on the malleability of administrative law to the political preferences of judges. Success is far from a sure thing, but it really could happen.

The larger point concerns the separation of powers system. It was designed to be, and is, resistant to fleeting passions and short-term power grabs. But only up to a point. If we look six years out rather than two, Republican dominance of the D.C. Circuit and the Supreme Court becomes almost certain.

In the end, it devolves to the American voter. If the electorate in 2020 re-elects Trump and a Republican Senate, they are quite likely to get a regulatory revolution, one way or another. But, in a democracy, would any other result be proper? Back-to-back presidential victories and multiple bouts of Senate control would surely constitute a mandate. It will then be cold comfort for liberals to recall H. L. Mencken’s definition of our form of government: “Democracy is the theory that the common people know what they want, and deserve to get it good and hard.”