Grok Lab: Sports Underdog Bias +resolution
Never fired. It was a forward test that produced no paper trade before it was archived.
0paper trades closed
—net, paper
—trades that won
What it tried — Grok Lab: Sports Underdog Bias +resolution
Forward test — Grok tried to fix the losing bot "Sports Underdog Bias" by adding the resolution_anchor smart signal. Trading live paper, judged on its FORWARD record vs the parent (no backtest — this signal can't be replayed). Bot's 21% hit rate on underdog YES (0.15-0.35) shows fighting calibrated crowd. resolution_anchor gates to buckets with >=0.6 historical win rate over 30 samples for real ed
The entry rule, as configured (sizing: flat, $5.0 a fill):
- closes within (h)
- 24
- resolution_anchor
- {'min_samples': 30, 'min_win_rate': 0.6}
- 24h volume at least
- 5000
- YES ask at most
- 0.35
- YES ask at least
- 0.15
Every version — Grok Lab: Sports Underdog Bias +resolution
| Version | Status | Snapshot | Change |
|---|---|---|---|
| v1 | active | 2026-06-11 | Forward-test fork (added resolution_anchor) of losing bot sports-underdog-bias |
Paper trades only — simulated fills at quoted prices, never money. The formula is archived, not deleted. Back to the Graveyard · What each signal means.