A prediction is only as good as the evidence behind it.
Why we’re starting with sources, uncertainty, and an open research notebook—not a leaderboard.
Read the essay ↗A public research notebookCan we model
Can we model
the Court’s
reasoning?
We’re exploring an open-source approach to Supreme Court prediction. Follow the evidence, the experiments, and the questions we haven’t answered yet.
Inside the research ↗THE RESEARCH QUESTIONFIG. 01
?
Currently in the lab
Building the source inventory
Snapshot · September 30, 202654Source metadata records
9Official opinion document links
Not yetPrediction accuracy measured
Notes from the lab
All entries ↗The public research notebook begins.
A home for project notes, evidence snapshots, and the small steps behind a larger research question.
Read the update ↗Keeping the answer out of the question.
A historical test can look impressive for the wrong reason. Here’s the evaluation rule we’re designing around.
Read the note ↗The work ahead
Show the process.
Earn the prediction.
A convincing model starts long before a confident answer. Here’s what has to come next.
Explore our approach ↗- 01Curate original, attributable sources
- 02Build and inspect the retrieval baseline
- 03Test without access to the answers
- 04Publish results, uncertainty, and failures