practice leaderboard 40 episodes / role / reference seed base 20260901 (published)
| # | Student | Best score | as Iran | as U.S. | Scored upload (UTC) | Uploads |
|---|---|---|---|---|---|---|
| 1 | 401107613 | 61.7 | 75.0 | 48.5 | 2026-08-07 11:38:00 UTC | 2 |
| 2 | 402111237 | 61.7 | 72.9 | 50.4 | 2026-08-09 07:06:06 UTC | 4 |
| [benchmark] MasterAgent | 56.3 | 65.7 | 46.9 | — | — | |
| 3 | 400104986 | 54.9 | 79.0 | 30.8 | 2026-08-08 14:28:12 UTC | 11 |
| [baseline] TitForTat | 53.8 | 69.7 | 37.9 | — | — | |
| [baseline] Hawk | 51.5 | 71.9 | 31.2 | — | — | |
| 4 | 403110636 | 49.9 | 63.4 | 36.5 | 2026-08-10 04:43:40 UTC | 2 |
| [baseline] Random | 24.7 | 30.7 | 18.6 | — | — | |
| [baseline] Dove | 20.3 | 25.7 | 14.9 | — | — | |
| [baseline] BayesianThreshold | 19.5 | 33.7 | 5.2 | — | — |
Score = mean raw utility ui against the reference pool (five scripted baselines + the instructor benchmark), both roles, common practice seeds. Best scored upload per team counts. Italic rows are fixed references, not submissions. The official ranking is a full team-vs-team round robin on fresh, undisclosed seeds — see Official results.