Will the Supreme Court rule by July 3, 2027 that Arizona can require documentary proof of citizenship from voters who register with the state form?

The key votes are Chief Justice Roberts and Justice Kavanaugh. In August 2024 they joined Justices Thomas, Alito and Gorsuch to let Arizona enforce the state-form proof requirement while the case continued, even though they declined to let the rest of the law take effect before the election. Voting to let a rule operate and then striking it down on the merits is possible, but justices rarely reverse their own interim judgment without a new argument that changes the picture.

Justice Barrett is the uncertain vote. In 2024 she sided with the three liberal justices to deny the application entirely. That vote may have been driven by the timing concern of changing election rules months before an election, which does not apply to a merits ruling in 2027. Her record on federal preemption favors reading federal statutes by their text, and the National Voter Registration Act's accept and use command speaks to the federal form, not a state's own form.

Chief Justice Roberts has institutional reasons to decide this well before the 2028 cycle and to write narrowly. The most natural narrow ruling for him is that Congress regulated the federal form and left the state form alone, which is a YES on our question even if the Court sides with the challengers on the 90-day removal issue. Our first version understated who is asking for that ruling: the State of Arizona brought its own petition alongside the RNC's, and on August 28 Arizona and Attorney General Kristin Mayes filed a merits brief in partial support of the petitioner on the state-form question.

The closest analogue is Arizona v. Inter Tribal Council in 2013, where the Court ruled 7-2 that the NVRA preempts Arizona's proof requirement for people using the federal form. Justice Scalia wrote that opinion and Roberts joined it. It cuts both ways: it shows the Court reads the NVRA seriously, but its reasoning rested on the federal form specifically, which leaves the state form open. That is why Arizona built its two-track system in the first place.

The second analogue is Beals v. Virginia Coalition for Immigrant Rights in October 2024, where the Court allowed Virginia to keep removing roughly 1,600 alleged noncitizens from its rolls inside the NVRA's 90-day quiet period, over the dissent of the three liberal justices. Together with the August 2024 Arizona order, the Court has granted emergency relief to the state on both questions now before it, as the RNC's petition emphasizes.

Where the analogies break: emergency orders are not merits rulings, and the consent decree issue is new. A 2018 agreement by a prior Arizona Secretary of State to accept state forms without documents is a separate barrier, and the Court could find the NVRA allows the requirement but the decree still binds. Most of the conservative majority has been skeptical of consent decrees that bind future officials, which limits that risk.

Start with the Court's overall reversal rate, about 70% for the cases it chooses to hear, and higher for the Ninth Circuit. That gives a prior of about 0.72 for the RNC on at least one question.

Adjust up for the August 2024 interim order, where five justices sided with Arizona on exactly this state-form question: plus 8 points. Adjust down for the consent decree as an independent ground to rule against Arizona: minus 4 points. Adjust down for a dismissal, a standing problem, or a split decision where the Court rules against Arizona on proof but for it on removals: minus 4 points. 0.72 plus 0.08 minus 0.08 equals 0.72.

Market comparison, corrected. The Kalshi contract requires Arizona to win on both questions. Our matching scenario, Arizona wins both, is 52 percent, so on the comparable outcome we are well below the market's last verified 69 cents, not above it. We are lower mainly because we give real weight to the challengers on the 90-day removal question, where the statute speaks of systematic removal programs and the Court's 2024 Virginia order was unexplained. The market has traded fewer than a hundred contracts, so we do not move toward it.

October 12 update, a correction rather than news. We start from version 1's 72 percent. Version 1's adjustments folded dismissal into a combined minus 4 points alongside split decisions, while its scenario table put 6 points on a dismissal, standing defect or no decision, partly because it treated the RNC and legislative leaders as the law's only defenders. That was wrong: the State of Arizona also petitioned for review, so the Court has a state party with an undisputed interest in its own law. We cut the dismissal scenario from 6 to 3 points: 72 plus 3 is 75 percent. Nothing else in the record moved before the December 8 argument.