Kalshi odds a week before each Supreme Court decision

How Accurate Are Prediction Markets on Supreme Court Cases? Kalshi’s Track Record

Kalshi’s Supreme Court markets picked the winning side 17 of 21 times a week before the decision.

Every settled Kalshi market on the outcome of a Supreme Court case, priced seven days and one day before the Court ruled. Decision dates are checked against the Court’s own opinion list.

Kalshi data and supremecourt.gov, checked Oct 12, 2026, 10:03 AM ET. Updated weekly.

By Arthur Angelo, PredictionScout Research · Updated Oct 12, 2026, 10:03 AM ET

17 of 21favorite right, 7 days before
18 of 21favorite right, 1 day before
12 of 13right when priced 80%+
0.110Brier score (lower is better)

Key findings

  • 17 of 21 right a week out. In settled Kalshi markets on how the Supreme Court would rule, the side priced above 50% seven days before the decision won 17 times (81%). One day before, it was 18 of 21.
  • Confident calls held up. When the market was at 80% or more for one side a week out, that side won 12 of 13 times.
  • Where the market was wrong: Consumer Financial Protection Bureau v. Community Financial Services Assn. (cfpb funding structure found unconstitutional: 60% a week before, outcome no); TikTok Inc. v. Garland (certiorari) (court agrees to hear the tiktok case before july 1, 2025: 38% a week before, outcome yes); FDA v. Wages and White Lion Investments, LLC (court rules against the fda: 54% a week before, outcome no); Watson v. Republican National Committee (court bars counting mail ballots received after election day: 80% a week before, outcome no).
  • Small sample. These are all 21 case-outcome markets on Kalshi that settled and had a price a week before the decision, from Apr 14, 2023 to Sep 21, 2026. Brier score 0.110 (0 is perfect; always guessing 50% scores 0.25).

Odds a week before each decision

Kalshi odds a week before each Supreme Court decisionKalshi odds a week before each Supreme Court decision024487296120Axon Enterpris v.: 92National Pork v.: 24.5Gonzalez v.: 12.5Glacier Northw v.: 87.5Students for F v.: 40.5Students for F v.: 59.0Biden v.: 25Consumer Finan v.: 59.5Moore v.: 7.0Loper Bright E v.: 94TikTok Inc. v.: 38TikTok Inc. v.: 30.0FDA v.: 54Trump v.: 5.5Louisiana v.: 93.6United States v.: 14.0Trump v.: 86.5Watson v.: 80Trump v.: 91.9West Virginia v.: 93.6Genalo v.: 4Axon Enterpris v.Glacier Northw v.Biden v.Loper Bright E v.FDA v.United States v.Trump v.Genalo v.Chance of YES 7 days before (%)%Kalshi; supremecourt.gov; PredictionScout
Kalshi’s implied chance of YES seven days before each decision. See the table for what YES meant and the outcome.

Every case market

“YES means” describes the market question in plain words; see the Kalshi market for exact rules.

DecidedCaseYES meansChance of YES, 7 days before1 day beforeOutcomeFavorite right?Contracts
Sep 21, 2026Genalo v. BlackCourt rules for Kenneth Genalo4%5%No✓8,074
Jun 30, 2026West Virginia v. B. P. J. / Little v. HecoxCourt upholds the state laws94%94%Yes✓30,741
Jun 30, 2026Trump v. Barbara (rehearing)Court grants rehearingNo27,142
Jun 29, 2026Trump v. SlaughterCourt rules for the President92%93%Yes✓58,364
Jun 29, 2026Watson v. Republican National CommitteeCourt bars counting mail ballots received after Election Day80%76%No✗164,491
Jun 25, 2026Trump v. MiotCourt rules for the administration (Haiti TPS)86%70%Yes✓31,591
Jun 18, 2026United States v. HemaniCourt rules for the United States14%16%No✓73,435
Apr 29, 2026Louisiana v. CallaisCourt rules for the State of Louisiana94%93%Yes✓35,676
Apr 7, 2025Trump v. J. G. G.Unanimous ruling for the government6%6%No✓282
Apr 2, 2025FDA v. Wages and White Lion Investments, LLCCourt rules against the FDA54%54%No✗49,562
Jan 17, 2025TikTok Inc. v. GarlandCourt rules in TikTok's favor30%18%No✓97,610
Jan 17, 2025TikTok Inc. v. Garland (certiorari)Court agrees to hear the TikTok case before July 1, 202538%75%Yes✗109,241
Jun 28, 2024Loper Bright Enterprises v. RaimondoCourt rules for the petitioners (Chevron weakened or overruled)94%92%Yes✓19,875
Jun 20, 2024Moore v. United StatesMandatory repatriation tax found unconstitutional7%6%No✓8,021
May 16, 2024Consumer Financial Protection Bureau v. Community Financial Services Assn.CFPB funding structure found unconstitutional60%60%No✗40,434
Jun 30, 2023Biden v. NebraskaInjunction on student-loan forgiveness lifted25%31%No✓40,686
Jun 29, 2023Students for Fair Admissions, Inc. v. President and Fellows of HarvardUniversal ban on race-conscious admissions59%62%Yes✓128,040
Jun 29, 2023Students for Fair Admissions, Inc. v. President and Fellows of HarvardBan on Harvard’s race-conscious admissions40%38%No✓69,223
Jun 1, 2023Glacier Northwest, Inc. v. TeamstersCourt rules federal labor law does not preempt the state tort claim88%87%Yes✓3,859
May 18, 2023Gonzalez v. Google LLCCourt rules Section 230 does not protect recommendation algorithms12%10%No✓4,450
May 11, 2023National Pork Producers Council v. RossCourt strikes down California's Proposition 1224%30%No✓19,573
Apr 14, 2023Axon Enterprise, Inc. v. FTCCourt holds district courts can hear the constitutional challenge92%92%Yes✓1,778

Not scored: 3 markets on the exact number of justices voting for one side (in Will exactly 9 justices vote for the petitioner in Donald J., Will exactly 9 justices vote for the petitioner in Louisiana, Will exactly 9 justices vote for the petitioner in Trump v. ). Most of those contracts traded thinly, so their prices are unreliable. They are not in the headline numbers. Cert-petition list markets (“which cases will the Court agree to hear”) are also excluded.

Download the data

Every case market (CSV)

One row per settled case-outcome market with the implied chance 7 and 1 days before the decision. Free to reuse under CC BY 4.0 with a link back to this page.

Embed this chart

Methodology

  • Markets. Settled Kalshi markets in its Supreme Court case series that ask how a specific case will come out (or whether the Court will take a specific case), from Kalshi’s public API. We exclude markets about appointments, retirements, timing and vote counts.
  • Decision date. Taken from the Court’s slip opinion lists for each term. Where a case is not on those lists (orders and emergency applications), we use the time Kalshi closed the market. Older Kalshi markets sometimes stayed open for months after a ruling, so we never use the close time when the Court’s date is available.
  • Price. Midpoint of bid and ask (spread 10¢ or less; otherwise the last trade) from the last hourly candle before 9:00 AM ET, 7 days and 1 day before the decision, falling back to daily candles up to three weeks back.
  • Neutrality. We report what the markets priced and what the Court decided. Nothing here comments on the merits of any case.
  • Limits. 21 markets is a small sample, and several traded only a few thousand contracts. Last run: Oct 12, 2026, 10:03 AM ET.

Cite this research

Suggested citation: PredictionScout. (2026, October 12). How Accurate Are Prediction Markets on Supreme Court Cases? Kalshi's Track Record. https://predictionscout.com/research/supreme-court-prediction-market-accuracy/

For a chart or news story: “Source: PredictionScout analysis of public exchange data (https://predictionscout.com/research/supreme-court-prediction-market-accuracy/)”

The data and charts on this page are free to reuse under CC BY 4.0 with a link back to this page. Market data: Kalshi. Decision dates: Supreme Court of the United States.

Data sources: prices and volumes come from the public APIs of Kalshi and Polymarket. All trademarks belong to their owners. PredictionScout is independent and not affiliated with any exchange. Nothing here is investment, betting, or legal advice.

Related: prediction market lawsuit tracker · accuracy scorecard