Judge-level disparity
Seventh Circuit · 2020–2025 · 1,591 decisions
Decision relief
11.3%156 of 1,380 known outcomes
Counts & definition
Any reversal, vacatur, remand, or comparable appellate relief. Original v14 classifications.
All decisions: 156 relief / 1,380 known; 21 unknown, excluded from the rate; 1,401 total decisions.
Routine: 28 relief / 732 known; 17 unknown, excluded from the rate; 749 total decisions.
Doctrine: 128 relief / 648 known; 4 unknown, excluded from the rate; 652 total decisions.
Sentence relief
12.2%91 of 748 known outcomes
Counts & definition
Relief tied specifically to the sentence or further sentencing proceedings. Original v14 classifications.
All decisions: 91 relief / 748 known; 653 unknown, excluded from the rate; 1,401 total decisions.
Routine: 14 relief / 388 known; 361 unknown, excluded from the rate; 749 total decisions.
Doctrine: 77 relief / 360 known; 292 unknown, excluded from the rate; 652 total decisions.
Judge comparisons
Decision relief · 1,375 decisionsOpen circles: before case-mix adjustment. Filled circles: after. Both use the same publication mix.
Observed outcomes & model estimates
Decision relief · 1,375 decisionsDispersion comparison
| Track | Dispersion ratio | p value |
|---|---|---|
| All decisions | 2.84 | < 0.001 |
| Routine | 1.23 | 0.212 |
| Doctrine | 1.35 | 0.172 |
How the comparison is made
- Appellate decisionsSeventh Circuit, 2020–2025
- Extracted factsOutcomes, issues & case characteristics
- Two publication tracksRoutine & Doctrine
- Case-mix adjustmentShared publication mix
- Judge comparisonsCases associated with each panel member
Analysis details
Original thesis v14 analysis. Judge comparisons use decision relief, not sentence relief: 1,375 decisions and 4,157 panel-associated rows. The relief summaries above use 1,401 tracked decisions.
Routine / nonprecedential weight: 52.9%. Doctrine / precedential weight: 47.1%. The displayed order is retained from the supplied snapshot. These are panel-associated outcomes, not individual votes.
The underlying track estimates are the frozen thesis decision-relief case-mix-standardized panel model. No sentence-relief judge ordering is available in this pilot. Composite intervals and rank bands are pilot approximations derived from the exported track-specific 95% intervals, not refitted posterior chains.
Case-review classifications have not changed these analytical results.
Relief requires a material change in the party’s legal position. Clerical corrections, duplicate orders, corrected captions, and non-substantive judgment modifications are not decision relief or sentence relief.