Eric M. Nakata
F++
50% — 16 of 32 rulings upheld, per qwen3.8-27b; in the top 83% of 312 graded judges
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.
43cases reviewed on appeal
32%reversed, in whole or part, of 41 merits decisions
Outcomes on appeal
affirmed 21 reversed 8 modified 7 reversed in part 5 writ denied 1 dismissed 1
Areas
By year: 2014 11 · 2016 10 · 2015 6 · 2022 5 · 2021 4 · 2013 4 · 2020 1 · 2018 1 · 2017 1 ·
Analysis
24reviewed cases analysed
101rulings read
2rulings where the model disagrees with the court
4cases with a verified missed authority
1appellate 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. Session CA4/2
- People v. Sherwood CA4/2
- People v. Lattin CA4/1
- People v. Abad CA4/1
- People v. Rebosio CA4/2
- People v. Session CA4/2
- People v. Riley CA4/1
- People v. Reyes CA4/2
- People v. Hernandez CA4/2
- People v. Steen CA4/2
- People v. Arebalos-Cabrera
- People v. Echavarria
- People v. Kennedy CA4/2
- People v. Gibson
- People v. Johnson CA4/2
- People v. Hardeman CA4/1
- People v. Reyes
- People v. Reyes
- People v. Marquez CA4/2
- People v. Love CA4/1
- People v. Hogue CA4/2
- People v. Cage CA4/1
- People v. Koch CA4/2
- People v. Allen CA4/2
- People v. Abad CA4/1