John M. Davis
C+
provisional — 67% — 6 of 9 rulings upheld, per qwen3.8-27b; would rank in the top 45% of 312 graded judges, but ten rulings are needed for a standing letter — 9 so far
On the assumption that the model is right — qwen3.8-27b reads the court's own disposition right 70% of the time, graded on a curve: A is the top fifth of the judges graded under this model, F the bottom fifth; ++ / + / − / −− within the band. How the grade is made.
33cases reviewed on appeal
24%reversed, in whole or part, of 33 merits decisions
Outcomes on appeal
affirmed 20 modified 5 reversed in part 4 reversed 4
Areas
By year: 2021 10 · 2020 6 · 2014 4 · 2025 3 · 2026 2 · 2022 2 · 2016 2 · 2024 1 · 2023 1 · 2015 1 · 2013 1 ·
Analysis
10reviewed cases analysed
29rulings read
0rulings where the model disagrees with the court
1cases with a verified missed authority
0appellate opinions called questionable
Counts of a model's readings across this judge's reviewed cases, each shown on its case page beside the court's disposition. Not a finding about the judge.
Cases
- People v. Padilla CA4/2
- People v. Zapata
- People v. Ruiz CA4/1
- People v. Winkle CA4/1
- People v. Seremak CA4/2
- People v. Scott CA4/3
- People v. Gomez CA4/2
- People v. Venegas CA4/2
- People v. Wright CA4/1
- People v. Gant CA4/2
- People v. Reddy CA4/2
- People v. Edgin CA4/2
- People v. Reddy CA4/2
- People v. Tshughuryan CA4/2
- People v. Velazquez CA4/1
- People v. Bernhard CA4/2
- People v. Garcia CA4/2
- People v. Ramirez CA4/2
- McLane v. GoPlus Corp. CA4/2
- People v. Coto CA4/2
- People v. Moss CA4/2
- People v. Dorsey CA4/2
- People v. Andrews CA4/2
- People v. Mendoza
- People v. Mendoza