Skip to main content
Independent · not a government site
A row of marble Corinthian columns along the facade of the Supreme Court Building.

81 terms · 9,420 decisions

Trends & patterns

How the Court's caseload, vote splits, and issue focus have shifted since 1946.

Photo: Daderot · Public domain · Wikimedia Commons · All photo credits

Charts

Cases decided per term

The docket has shrunk by roughly two thirds. The Court decided 197 cases in the 1967 term and 74 in 2025.

5-4 decisions over time

Bare majorities peaked at 44 in the 1986 term. Since 2005 the Court has averaged 12.0 a term — fewer in number, out of a much smaller docket.

Majority-size distribution

Unanimity is the single most common outcome: 9-0 accounts for 2,726 of 9,420 decisions, more than any divided split.

Issue areas by decade

What the Court spends its docket on has shifted by decade — each bar is one decade’s decisions, split by issue area.

Petitioner win rate

The petitioner usually wins — about 64% across the period. The Court tends to grant review when it suspects error below, and the line sits above the even split for most of it.

Decision direction by term

Coded direction swings by era rather than trending. Across the whole period it is close to even: 4,442 conservative against 4,629 liberal.

Charts through the 2023 term are computed from the Supreme Court Database (SCDB), 2024 Release 01. Later terms count slip opinions. Issue area, direction, and petitioner-win charts stay on the SCDB coding, which does not cover those terms.

What the charts show

The caseload line is the headline: an average of 146 signed decisions a term across 1946–1985, peaking at 197 in 1967, against an average of 67 over the last ten terms. The 2025 term produced 74 decisions.

Bare majorities moved the other way: 5–4 splits were 13.1% of decisions in 1946–1985 and 17.3% from 2005 on. Across the whole dataset 2,726 decisions were unanimous and 1,408 were 5–4.

Issue-area and direction codes come from SCDB’s coding conventions and involve editorial judgement — they are most useful for longitudinal comparison, not for characterising any single case. See Methodology & sources.