The Supreme Court — the record
The model's full walk-forward backtest, 1970–2025 (54,465 justice-votes), and how it stacks up against the published academic benchmark and every other forecaster we track. ← Back to the current court
How well it forecasts
Scored strictly walk-forward over 1970–2025. The honest bar is the Court's strong habit of siding with the petitioner (it reverses more than it affirms) — so "petitioner always wins" is a tough baseline, and it has only gotten tougher as recent benches grew more lopsided. The durable edge is at the justice level.
Per-justice vote accuracy vs the baseline, by term
The model holds a gap above the "petitioner always wins" line across the era. The gap narrows recently because the baseline itself climbs — modern terms break the petitioner's way more often, leaving less room to beat.
| By issue area | Model | Base | n |
|---|---|---|---|
| Criminal Procedure | 68.6% | 68.7% | 1,503 |
| Economic Activity | 63.8% | 63.5% | 1,119 |
| Civil Rights | 64.1% | 64.3% | 1,101 |
| Judicial Power | 61.5% | 61.6% | 833 |
| First Amendment | 66.6% | 67.0% | 443 |
| Federalism | 60.7% | 64.5% | 290 |
| Due Process | 70.2% | 69.8% | 275 |
| Unions | 57.9% | 59.5% | 195 |
| Federal Taxation | 67.8% | 55.5% | 146 |
| Privacy | 68.1% | 69.0% | 116 |
| Attorneys | 69.3% | 65.9% | 88 |
| Interstate Relations | 59.5% | 51.4% | 74 |
All models, compared
Every rule that turns information into a prediction is a model. That includes VotePredictor, the FantasySCOTUS crowd, the published {Marshall}+ method, and the simple "petitioner always wins" model. We group all of them here. Scores are only ranked within identical evaluation samples; different periods, cases, or prediction units are shown together but not treated as a head-to-head race.
Long-run models · same 5,914 cases · 1971–2023
VotePredictor and our faithful reproduction of {Marshall}+ are trained walk-forward. The petitioner-always-wins model is a deterministic base-rate model. All three are scored on the same underlying cases below.
VotePredictor SCOTUS vs {Marshall}+ (reproduction)
same cases · 1970–2025 · 6,184 cases| Metric | VotePredictor SCOTUS | {Marshall}+ | Petitioner-always-wins model |
|---|---|---|---|
| Justice-vote accuracy | 67.3% | 65.5% | 60.6% |
| Justice-vote Brier (lower better) | 0.208 | 0.216 | — |
| Case accuracy (vote aggregation) | 66.9% | 66.2% | — |
| Justice-vote accuracy by decade | 1970s | 1980s | 1990s | 2000s | 2010s | 2020s |
|---|---|---|---|---|---|---|
| VotePredictor SCOTUS | 68.1% | 67.5% | 64.8% | 67.6% | 66.9% | 69.5% |
| {Marshall}+ | 66.6% | 66.8% | 63.4% | 65.1% | 63.3% | 67.1% |
| Petitioner-always-wins model | 61.2% | 58.0% | 57.2% | 64.2% | 63.3% | 64.5% |
{Marshall}+is our faithful reproduction of the Katz–Bommarito–Blackman (2017) method — extremely-randomized trees on the full set of raw SCDB codes — since their actual per-case predictions aren't published. Their reported headline was 71.9% justice / 70.2% case over 1816–2015; that span has a lower base rate than the modern window here, so it's context rather than a like-for-like target. On identical modern cases, our model edges the classic approach.
Crowd model · recovered same-case sample · 2019–2024
FantasySCOTUS is a crowd-aggregation model, not a separate category. We recovered its public pages from the Internet Archive and kept only snapshots captured before the decision. The cleanest comparison has 39 direct case-outcome crowds from October Terms 2019–2024. Archive coverage is incomplete, so this is an audited sample, not directly comparable to the long-run table above.
| Same archived cases | n | accuracy | Brier |
|---|---|---|---|
| FantasySCOTUS crowd model | 39 | 69.2% | 0.225 |
| VotePredictor SCOTUS | 39 | 61.5% | 0.237 |
A strictly forward blend—its crowd weight chosen using earlier terms only—scores 72.0% accuracy and 0.220 Brier on 25 later cases, versus 68.0% / 0.234for the crowd and 68.0% / 0.234for our model on those same rows. That result is promising but small-sample. Another 53 cases have only per-justice crowd probabilities; they remain labeled as a separate reconstructed series rather than being passed off as direct case forecasts.
The model zoo — every forecaster, ranked
Rather than pick one design, we build them all and let the data rank them — forecasting October Term 2025 cases after oral argument. Our model is VotePredictor SCOTUS (a per-justice model on the Supreme Court Database — separate from the VotePredictor Elections model) with oral-argument analysis added — the structural-only variant is scored below but no longer served as a live row, because on its own its live probability was a calibrated base rate — alongside the FantasySCOTUS crowd, prediction markets, and a consensus blend. Two honest takeaways: the informed crowd wins, and analyzing the oral argument is the strongest single signal we add — historically beating our structural model on its own.
| Forecaster | Brier (lower = better) | Accuracy (base rate 85.2%) | n |
|---|---|---|---|
| FantasySCOTUS crowd | 0.122 | 96.3% | 27 |
| Consensus (model + crowd + market)not independent | 0.126 | 88.9% | 27 |
| VotePredictor SCOTUS (after oral argument) | 0.149 | 85.2% | 27 |
| Prediction marketstoo few cases to rank | 0.011 | 100.0% | 1 |
| "Petitioner always wins" base rate | — | 85.2% | 27 |
Does analyzing the oral argument help? (2013–2023, walk-forward)
On 641 argued cases with transcripts. Honest reading: the structural model alone carries no case-level signal on these cases (its ranking of them is no better than a coin flip), so its probability is a calibrated base rate; the oral-argument signal (how hard the justices grill each side) is where the information is, and the two together score best on Brier. None beats the lopsided base rate on raw accuracy.
| Model | Accuracy | Brier |
|---|---|---|
| Structural features only (retired as a live row) | 65.8% | 0.226 |
| Oral argument only | 63.7% | 0.239 |
| VotePredictor SCOTUS (structural + oral) | 68.2% | 0.212 |
| baseline | 67.6% | — |
| Case | VotePredictor SCOTUS | crowd | market | consensus | result |
|---|---|---|---|---|---|
| learning resources v. trump Roberts 76Thomas 76Alito 69Sotomayor 63Kagan 66Gorsuch 76Kavanaugh 76Barrett 76Jackson 63 | 70% | 57% | — | 63% | Petitioner |
| robinson v. callais Roberts 69Thomas 69Alito 69Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 69Barrett 76Jackson 76 | 70% | 41% | — | 55% | Respondent |
| chiles v. salazar Roberts 76Thomas 80Alito 85Sotomayor 41Kagan 56Gorsuch 85Kavanaugh 77Barrett 85Jackson 41 | 74% | 73% | — | 73% | Petitioner |
| wolford v. lopez Roberts 85Thomas 83Alito 85Sotomayor 80Kagan 80Gorsuch 85Kavanaugh 85Barrett 85Jackson 80 | 74% | 64% | — | 68% | Petitioner |
| noem v. al otro lado Roberts 76Thomas 80Alito 80Sotomayor 27Kagan 39Gorsuch 85Kavanaugh 80Barrett 80Jackson 27 | 74% | 68% | — | 71% | Petitioner |
| cox communications v. sony music Roberts 65Thomas 61Alito 65Sotomayor 72Kagan 76Gorsuch 76Kavanaugh 76Barrett 65Jackson 65 | 74% | 94% | — | 85% | Petitioner |
| fcc v. at&t Roberts 76Thomas 76Alito 76Sotomayor 76Kagan 76Gorsuch 76Kavanaugh 76Barrett 76Jackson 76 | 70% | 70% | — | 70% | Petitioner |
| chevron usa v. plaquemines parish Roberts 65Thomas 56Sotomayor 56Kagan 52Gorsuch 65Kavanaugh 65Barrett 65Jackson 61 | 65% | 72% | — | 69% | Petitioner |
| monsanto v. durnell Roberts 65Thomas 65Alito 65Sotomayor 59Kagan 65Gorsuch 61Kavanaugh 65Barrett 65Jackson 52 | 74% | 52% | — | 61% | Petitioner |
| hikma pharmaceuticals v. amarin Roberts 72Thomas 64Alito 76Sotomayor 65Kagan 80Gorsuch 80Kavanaugh 80Barrett 80Jackson 80 | 74% | 93% | — | 85% | Petitioner |
| exxon mobil v. corporacion cimex Roberts 76Thomas 65Alito 80Sotomayor 65Kagan 65Gorsuch 85Kavanaugh 77Barrett 76Jackson 65 | 74% | 61% | — | 67% | Petitioner |
| bost v. illinois state board of elections Roberts 85Thomas 76Alito 85Sotomayor 65Kagan 76Gorsuch 85Kavanaugh 85Barrett 77Jackson 72 | 74% | 79% | — | 77% | Petitioner |
| villarreal v. texas Roberts 65Thomas 43Alito 52Sotomayor 80Kagan 72Gorsuch 65Kavanaugh 52Barrett 61Jackson 65 | 74% | 16% | — | 44% | Respondent |
| barrett v. united states Roberts 41Thomas 43Alito 34Sotomayor 59Kagan 59Gorsuch 65Kavanaugh 43Barrett 41Jackson 65 | 57% | 76% | — | 68% | Petitioner |
| bowe v. united states Roberts 61Thomas 52Alito 41Sotomayor 77Kagan 80Gorsuch 59Kavanaugh 72Barrett 43Jackson 85 | 74% | 61% | — | 67% | Petitioner |
| usps v. konan Roberts 65Thomas 52Alito 65Sotomayor 39Kagan 34Gorsuch 50Kavanaugh 65Barrett 61Jackson 41 | 61% | 55% | — | 58% | Petitioner |
| olivier v. city of brandon Roberts 76Thomas 56Alito 65Sotomayor 72Kagan 76Gorsuch 65Kavanaugh 76Barrett 65Jackson 77 | 74% | 88% | — | 82% | Petitioner |
| first choice women's resource centers v. platkin Roberts 85Thomas 85Alito 85Sotomayor 65Kagan 85Gorsuch 85Kavanaugh 85Barrett 85Jackson 65 | 74% | 94% | — | 85% | Petitioner |
| hencely v. fluor Roberts 64Thomas 65Alito 52Sotomayor 72Kagan 76Gorsuch 76Kavanaugh 52Barrett 65Jackson 61 | 74% | 64% | — | 68% | Petitioner |
| landor v. louisiana dept. of corrections Roberts 72Thomas 56Alito 64Sotomayor 85Kagan 85Gorsuch 64Kavanaugh 76Barrett 77Jackson 80 | 74% | 48% | — | 60% | Respondent |
| galette v. new jersey transit Roberts 50Thomas 50Alito 43Sotomayor 65Kagan 65Gorsuch 61Kavanaugh 76Barrett 61Jackson 80 | 74% | 51% | — | 61% | Petitioner |
| t.m. v. university of maryland medical system Roberts 64Thomas 52Alito 52Sotomayor 65Kagan 61Gorsuch 65Kavanaugh 72Barrett 65Jackson 64 | 74% | 65% | — | 69% | Respondent |
| cisco systems v. doe i Roberts 80Thomas 85Alito 87Sotomayor 65Kagan 77Gorsuch 87Kavanaugh 85Barrett 85Jackson 76 | 74% | 74% | — | 74% | Petitioner |
| havana docks v. royal caribbean Roberts 65Thomas 65Alito 72Sotomayor 65Kagan 65Gorsuch 80Kavanaugh 65Barrett 76Jackson 76 | 74% | 60% | — | 66% | Petitioner |
| trump v. slaughter Roberts 64Thomas 85Alito 85Sotomayor 27Kagan 27Gorsuch 85Kavanaugh 77Barrett 52Jackson 27 | 74% | 63% | 89% | 75% | Petitioner |
| west virginia v. b.p.j. Roberts 77Thomas 76Alito 76Sotomayor 41Kagan 52Gorsuch 43Kavanaugh 64Barrett 56Jackson 30 | 70% | 71% | — | 71% | Petitioner |
| little v. hecox Roberts 87Thomas 85Alito 98Sotomayor 61Kagan 67Gorsuch 87Kavanaugh 87Barrett 85Jackson 61 | 74% | 68% | — | 71% | Petitioner |
| riseandshine corporation v. pepsico, inc. Roberts 69Thomas 69Alito 69Sotomayor 76Kagan 76Gorsuch 71Kavanaugh 69Barrett 69Jackson 76 | 70% | — | — | 70% | pending |
| guerrero v. johnson Roberts 76Thomas 76Alito 76Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 76Barrett 76Jackson 66 | 70% | — | — | 70% | pending |
| republican national committee v. mi familia vota Roberts 69Thomas 76Alito 76Sotomayor 53Kagan 54Gorsuch 69Kavanaugh 69Barrett 76Jackson 54 | 70% | — | — | 70% | pending |
| montoya palacios v. liggins Roberts 54Thomas 54Alito 52Sotomayor 69Kagan 63Gorsuch 59Kavanaugh 54Barrett 54Jackson 69 | 69% | — | — | 69% | pending |
| apple inc. v. epic games, inc. Roberts 69Thomas 69Alito 69Sotomayor 63Kagan 69Gorsuch 69Kavanaugh 69Barrett 69Jackson 58 | 70% | — | — | 70% | pending |
| hoffmann v. wbi energy transmission, inc. Roberts 53Thomas 53Alito 54Sotomayor 54Kagan 54Gorsuch 54Kavanaugh 53Barrett 54Jackson 66 | 66% | — | — | 66% | pending |
| suncor energy inc. v. county commissioners of boulder county Roberts 76Thomas 76Sotomayor 54Kagan 54Gorsuch 76Kavanaugh 76Barrett 76Jackson 54 | 70% | — | — | 70% | pending |
| crowther v. board of regents of the university system of georgia Roberts 44Thomas 44Alito 54Sotomayor 54Kagan 54Gorsuch 54Kavanaugh 58Barrett 54Jackson 54 | 63% | — | — | 63% | pending |
| viramontes v. cook county Roberts 69Thomas 69Alito 76Sotomayor 54Kagan 54Gorsuch 69Kavanaugh 69Barrett 69Jackson 58 | 70% | — | — | 70% | pending |
| jouppi v. alaska Roberts 58Thomas 53Alito 52Sotomayor 76Kagan 76Gorsuch 58Kavanaugh 63Barrett 54Jackson 69 | 70% | — | — | 70% | pending |
| younge v. fulton judicial circuit district attorney's office, georgia Roberts 69Thomas 54Alito 58Sotomayor 59Kagan 58Gorsuch 58Kavanaugh 66Barrett 59Jackson 58 | 70% | — | — | 70% | pending |
| nielsen v. watanabe Roberts 69Thomas 69Alito 69Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 69Barrett 69Jackson 58 | 70% | — | — | 70% | pending |
| salazar v. paramount global Roberts 54Thomas 54Alito 54Sotomayor 63Kagan 63Gorsuch 54Kavanaugh 54Barrett 58Jackson 69 | 69% | — | — | 69% | pending |
| anderson v. intel corporation investment policy committee Roberts 54Thomas 54Alito 54Sotomayor 69Kagan 69Gorsuch 54Kavanaugh 54Barrett 54Jackson 69 | 70% | — | — | 70% | pending |
| beaird v. united states Roberts 58Thomas 53Alito 53Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 58Barrett 54Jackson 69 | 70% | — | — | 70% | pending |
| department of the air force v. prutehi guahan Roberts 63Thomas 69Alito 69Sotomayor 54Kagan 54Gorsuch 63Kavanaugh 63Barrett 59Jackson 54 | 70% | — | — | 70% | pending |
| st. mary catholic parish v. roy Roberts 76Thomas 76Alito 76Sotomayor 66Kagan 69Gorsuch 76Kavanaugh 76Barrett 76Jackson 69 | 70% | — | — | 70% | pending |
| maxwell v. dinis Roberts 54Thomas 53Alito 44Sotomayor 69Kagan 63Gorsuch 54Kavanaugh 58Barrett 54Jackson 69 | 68% | — | — | 68% | pending |
| kian v. florida Roberts 58Thomas 52Alito 44Sotomayor 76Kagan 76Gorsuch 58Kavanaugh 63Barrett 54Jackson 76 | 70% | — | — | 70% | pending |
| johnson v. united states congress Roberts 69Thomas 58Alito 63Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 69Barrett 69Jackson 69 | 70% | — | — | 70% | pending |
| international partners for ethical care, inc. v. ferguson Roberts 76Thomas 71Alito 76Sotomayor 76Kagan 76Gorsuch 69Kavanaugh 76Barrett 76Jackson 76 | 70% | — | — | 70% | pending |
| wassily v. blanche Roberts 54Thomas 44Alito 44Sotomayor 69Kagan 69Gorsuch 58Kavanaugh 54Barrett 54Jackson 69 | 68% | — | — | 68% | pending |
| genalo v. black Roberts 69Thomas 76Alito 76Sotomayor 52Kagan 54Gorsuch 69Kavanaugh 69Barrett 69Jackson 54 | 70% | — | — | 70% | pending |
| grand v. city of university heights Roberts 76Thomas 69Alito 76Sotomayor 69Kagan 76Gorsuch 76Kavanaugh 76Barrett 76Jackson 76 | 70% | — | — | 70% | pending |
| department of labor v. sun valley orchards, llc Roberts 66Thomas 54Alito 54Sotomayor 63Kagan 66Gorsuch 58Kavanaugh 69Barrett 58Jackson 63 | 70% | — | — | 70% | pending |
| department of homeland security v. d.v.d. Roberts 71Thomas 76Alito 76Sotomayor 54Kagan 58Gorsuch 69Kavanaugh 76Barrett 69Jackson 58 | 70% | — | — | 70% | pending |
| missionaries of saint john the baptist, inc. v. frederic Roberts 76Thomas 76Alito 76Sotomayor 76Kagan 76Gorsuch 69Kavanaugh 76Barrett 76Jackson 76 | 70% | — | — | 70% | pending |
| marschner v. marschner Roberts 69Thomas 63Alito 63Sotomayor 69Kagan 69Gorsuch 69Kavanaugh 69Barrett 69Jackson 69 | 70% | — | — | 70% | pending |
| rhoney v. barbosa da cunha Roberts 69Thomas 76Alito 76Sotomayor 54Kagan 58Gorsuch 69Kavanaugh 69Barrett 69Jackson 58 | 70% | — | — | 70% | pending |
Cells show each forecaster's P(petitioner wins); green/red= right/wrong on decided cases. Under each case are the model's per-justice predictions — each justice's % chance of siding with the petitioner (name colored by appointing party, D/R; hover for detail). Forecasts are made after oral argument only. VotePredictor SCOTUS is our per-justice model on the Supreme Court Database plus the oral-argument signal, bench-wide and per justice (which justice grilled which side), read against how long each side held the podium — a separate model from VotePredictor Elections. The structural-only variant is no longer served as a live row: its live probability was a calibrated base rate (see the table above). Consensus blends model + crowd + market, weighted by each source's track record on this term's decided cases — it ingests the others, so it can't be benchmarked against them. Crowd = FantasySCOTUS; markets = Polymarket / Kalshi.
When the lower court builds on eroded precedent
A pattern we found while testing whether cited case law predicts reversal: when the court below leaned on a precedent the Supreme Court had already overruled, the Court reversed at a strikingly higher rate. It's a legible tell — the ground was visibly cracked before the decision — even though it's too rare to sharpen the model itself.
| Case (term) | Leaned on a precedent that was already overruled | Outcome |
|---|---|---|
| RIVERS v. GUERRERO (2024) | Teague v. Lane — overruled 2021 by Edwards v. Vannoy | Petitioner lost |
| CAMERON v. EMW WOMEN’S SURGICAL CENTER, P.S.C. (2021) | Apodaca v. Oregon — overruled 2020 by Ramos v. Louisiana | Petitioner won |
| DEVILLIER v. TEXAS (2023) | Williamson County Regional Planning Comm'n v. Hamilton Bank of Johnson City — overruled 2019 by Knick v. Township of Scott, Pennsylvania | Petitioner won |
| EDWARDS v. VANNOY, WARDEN (2020) | Harris v. United States — overruled 2013 by Alleyne v. United States | Petitioner lost |
| TOWN OF CHESTER v. LAROE ESTATES (2016) | McConnell v. Federal Election Commission — overruled 2010 by Citizens United v. FEC | Petitioner won |
| THOMPSON v. HEBDON (2019) | McConnell v. Federal Election Commission — overruled 2010 by Citizens United v. FEC | Petitioner won |
| MCCULLEN v. COAKLEY (2013) | Austin v. Michigan Chamber of Commerce — overruled 2010 by Citizens United v. FEC | Petitioner won |
| DARIN RYBURN, et al., PETITIONERS v. GEORGE R. HUFF, et al. (2011) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| CHADRIN LEE MULLENIX, PETITIONER v. BEATRICE LUNA, INDIVIDUALLY AND AS REPRESENTATIVE OF THE ESTATE OF ISRAEL LEIJA, JR., et al. (2015) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| WOOD v. MOSS (2013) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| LOS ANGELES COUNTY, CALIFORNIA, PETITIONER v. CRAIG ARTHUR HUMPHRIES et al. (2010) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| CHARLES A. REHBERG, PETITIONER v. JAMES P. PAULK (2011) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner lost |
| HERNANDEZ v. MESA (2016) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| JOHN D. ASHCROFT, PETITIONER v. ABDULLAH AL-KIDD (2010) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
| STEVE A. FILARSKY, PETITIONER v. NICHOLAS B. DELIA. (2011) | Saucier v. Katz — overruled 2009 by Pearson v. Callahan | Petitioner won |
"Recently-eroded" = overruled within 20 years before the case — and recency is the whole tell: leaning on a precedent overruled longago tracks the baseline, not this. Whether a precedent had been overruled is public before the Court rules, so it's a fair pre-decision signal. VotePredictor SCOTUS still doesn't use it as an input, for two reasons we measured: the model already reads these as reversal-prone from the case's posture (it calls them right ~88% of the time without this feature), and they're rare — 72 of 1169cases — too few to move the overall forecast either way. Shown here as an honest pattern, not a model feature. "Petitioner won" = the Court sided with the side that appealed, which is a reversal in most but not all cases.
How this works
A gradient-boosted model predicts, for every justice on a case, the probability they side with the petitioner, using only pre-decision features (issue area, jurisdiction, the lower court's ruling and its ideological direction, cert reason, case origin) plus that justice's own prior record and the Court's composition. A justice with no record yet starts from a prior fixed at appointment (their own score as a circuit judge, else the appointing president's), which fades as their votes come in. A separate case-level model produces the calibrated win probability. Both are trained walk-forward — to forecast a term, only earlier terms are used — so the accuracy above is out-of-sample. Vote data: Supreme Court Database(Spaeth et al.), the standard source for this work. "Ideology" in the profiles is the share of conservative votes, a descriptive proxy rather than a Martin–Quinn ideal point.