Nightly self-improvement · slate 2026-09-26
Generated 2026-09-27T05:00:00.010381-04:00 ET from official finals (mlb-api). Every learning stage is listed with what it did. "Blocked" means it wanted to act and a safety rule stopped it; "nothing to learn" means it had no eligible evidence.
Slate
2026-09-26
status completed
Winners
7/13
Brier 0.2600
Learned or changed
0
of 5 stages
Blocked
0
wanted to act and could not
Release
v4.31.0
DRIFT
What each learning stage did
| Stage | Outcome | Detail |
|---|---|---|
| KPI settlement | failed | did not run |
| Bullpen usage model | nothing to learn | No pregame bullpen forecasts to settle. Refit: not attempted. |
| Adaptive parameter proposal (Luna) | nothing to learn | candidate_registered: |
| Auto-apply | nothing to learn | no_promotable_candidate |
| Code agent | nothing to learn | disabled_by_config: code changes are made on the development host, never in the container. |
Games
| Game | Predicted | Actual | Home win prob | Result | Main driver | Starters |
|---|---|---|---|---|---|---|
| ARI@SD | 5-4 | 7-10 | 0.466 | miss | Run environment far higher than projected (under-dispersed) | Taylor Clarke: 18 proj / 2 real outs; Walker Buehler: 14 proj / 11 real outs |
| ATL@MIA | 5-4 | 8-3 | 0.470 | hit | Brent Suter: 15 proj / 9 real outs; Jack Ralston: 15 proj / 4 real outs | |
| CIN@TOR | 3-5 | 5-1 | 0.620 | miss | Starting-pitcher projection was wrong | Rhett Lowder: 12 proj / 18 real outs; Trey Yesavage: 15 proj / 6 real outs |
| CLE@KC | 6-5 | 11-5 | 0.440 | hit | Tanner Bibee: 16 proj / 18 real outs; Michael Wacha: 19 proj / 19 real outs | |
| COL@CHW | 4-9 | 9-6 | 0.698 | miss | Backed the wrong side with real conviction | Jose Quintana: 12 proj / 9 real outs; Davis Martin: 15 proj / 2 real outs |
| HOU@OAK | 6-5 | 13-2 | 0.477 | hit | Hayden Wesneski: 16 proj / 21 real outs; Jack Perkins: 14 proj / 15 real outs | |
| LAA@SEA | 4-5 | 4-8 | 0.518 | hit | Ryan Johnson: 16 proj / 15 real outs; Kade Anderson: 16 proj / 10 real outs | |
| LAD@SF | 4-2 | 4-3 | 0.447 | hit | Blake Snell: 17 proj / 12 real outs; Matt Wilkinson: 14 proj / 9 real outs | |
| NYM@WAS | 6-7 | 7-1 | 0.506 | miss | Starting-pitcher projection was wrong | Jonah Tong: 15 proj / 15 real outs; Connelly Early: 17 proj / 9 real outs |
| PIT@DET | 6-4 | 3-4 | 0.487 | miss | Starting-pitcher projection was wrong | Kirby Yates: 15 proj / 5 real outs; Justin Verlander: 11 proj / 16 real outs |
| STL@MIL | 4-5 | 2-3 | 0.535 | hit | Quinn Mathews: 18 proj / 15 real outs; Dustin May: 13 proj / 9 real outs | |
| TB@PHI | 4-2 | 12-1 | 0.392 | hit | Griffin Jax: 15 proj / 9 real outs; Aaron Nola: 18 proj / 3 real outs | |
| TEX@MIN | 4-5 | 6-2 | 0.510 | miss | Starting-pitcher projection was wrong | Nathan Eovaldi: 15 proj / 13 real outs; Bailey Ober: 15 proj / 15 real outs |
Day over day
Rolling evidence window against the prior night. Associations, not causal proof.
| Metric | Prior | Current | Change | Result |
|---|---|---|---|---|
| brier_score | 0.2407 | 0.2413 | 0.0006 | regressed |
| accuracy | 0.5882 | 0.5932 | 0.0050 | improved |
| team_run_mae | 2.5168 | 2.6314 | 0.1146 | regressed |
| total_run_mae | 3.6387 | 3.8559 | 0.2172 | regressed |
| calibration_error | 0.0736 | 0.0876 | 0.0140 | regressed |
| starter_er_mae | 1.7400 | 1.7895 | 0.0495 | regressed |
| starter_outs_mae | 3.6480 | 3.9123 | 0.2643 | regressed |
Calibration produced for the next slate
| Metric | Scale used | Measured | Rows | Status |
|---|---|---|---|---|
| No calibration produced. | ||||
Luna KPI candidates and their promotion gate
Promotion needs 400+ prospective rows over 5+ slates after the proposal, with a gain interval above zero. Only verified pregame forecasts count; local replays never do.
| KPI | Status | Target | Proposed | Rows | Slates | Gain interval | Progress |
|---|---|---|---|---|---|---|---|
| luna_pitcher_release_speed_vs_righthanded_starters | shadow | SO | 20260923 | 330/400 | 5/5 | [-0.0007, 0.0007] | |
| luna_pitcher_barrel_rate_vs_rhh_righthanded_starters | shadow | HR | 20260923 | 330/400 | 5/5 | [-0.0004, 0.0049] | |
| luna_batter_contact_rate_vs_breaking_righthanded_starters | shadow | H | 20260923 | 330/400 | 5/5 | [-0.0045, 0.0044] | |
| luna_batter_contact_rate_vs_breaking_left_handed_starters_for_strikeouts | shadow | SO | 20260924 | 122/400 | 4/5 | [-0.0027, 0.0021] | |
| luna_pitcher_contact_xwoba_allowed_vs_righthanded_starters_for_strikeouts | shadow | SO | 20260924 | 122/400 | 4/5 | [-0.0001, 0.0000] | |
| luna_pitcher_whiff_rate_breaking_vs_left_handed_hitters_for_hits | shadow | H | 20260924 | 122/400 | 4/5 | [-0.0056, -0.0011] | |
| luna_pitcher_home_run_rate_vs_righthanded_starters | shadow | HR | 20260925 | 122/400 | 4/5 | [-0.0010, 0.0056] | |
| luna_batter_ground_ball_rate_vs_fastball_vs_lhp_for_home_runs | shadow | HR | 20260926 | 86/400 | 3/5 | [-0.0001, -0.0000] | |
| luna_pitcher_whiff_rate_vs_right_handed_starters_for_batter_strikeouts | shadow | SO | 20260929 | 51/400 | 2/5 | [–, –] | |
| luna_batter_whiff_rate_vs_breaking_right_handed_starters_for_hits | shadow | H | 20260929 | 51/400 | 2/5 | [–, –] | |
| luna_batter_walk_rate_vs_righthanded_pitchers_for_home_runs | retired | HR | 20260924 | 0/400 | 0/5 | [–, –] |
Operator actions the cycle asked for (0)
Changes to code or a model recipe are never made by the nightly cycle. These are the measured findings that need one.
| Priority | Owner | Finding | Action |
|---|---|---|---|
| No operator actions requested. | |||
Luna post-mortem
No Luna post-mortem this night.