Skip to content

PR #572: Season blend 90/10 (divisor 80→55) + MIN_PROB 0.57→0.60 #441

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jaayslaughter-cpu merged 1 commit into from
May 15, 2026
+2 −2
Merged

PR #572: Season blend 90/10 (divisor 80→55) + MIN_PROB 0.57→0.60 #441

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions tasklets.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -292,7 +292,7 @@ def ping(*a, **kw): return False
# raise the bar from 52% without losing too many qualifying picks.
# At 0.55 with correct multipliers: 2-leg PP needs 55%^2 * 3 - 1 = -9.2% → still
# needs higher prob per leg, but combined_ev gate (+3%) now does the real work.
MIN_PROB = 0.57
MIN_PROB = 0.60 # PR #572: May 15 schedule — raised from 0.57
Copy link
Copy Markdown

@gemini-code-assist gemini-code-assist Bot May 15, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The update to MIN_PROB to 0.60 is not reflected in the duplicate file api/tasklets.py (line 288), which still uses 0.57. This inconsistency can lead to divergent behavior across the system. Additionally, the surrounding comments (lines 291 and 293) still refer to a value of 0.55 and should be updated to maintain documentation accuracy.


# Per-prop MIN_PROB overrides — loaded from data/calibration_params.json
# Format: {"hits": 0.62, "total_bases": 0.62, ...}
Expand Down Expand Up @@ -2436,7 +2436,7 @@ def _prewarm_career_stats():
# blend is consistent across all data paths and XGBoost training rows.
_season_start = OPENING_DAY # 2026-03-26
_days_played = max(0, (_today_pt() - _season_start).days)
_season_weight_2026 = round(min(1.0, _days_played / 80.0), 3)
_season_weight_2026 = round(min(1.0, _days_played / 55.0), 3) # PR #572: 90/10 by day 49 (May 15)
Copy link
Copy Markdown

@gemini-code-assist gemini-code-assist Bot May 15, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The season weight divisor update to 55.0 is missing from the duplicate definition in api/tasklets.py (line 2065). Additionally, the comment on line 2434 (context) still refers to game ~80, and the comment on this line (90/10) is slightly imprecise compared to the actual math (89.1%) and the PR description (89/11).

Suggested change
_season_weight_2026 = round(min(1.0, _days_played / 55.0), 3) # PR #572: 90/10 by day 49 (May 15)
_season_weight_2026 = round(min(1.0, _days_played / 55.0), 3) # PR #572: 89/11 by day 49 (May 15)

_season_weight_2025 = round(1.0 - _season_weight_2026, 3)

def _is_pre_match(game_id: str) -> bool:
Expand Down