Brier score, Platt scaling parameters, and confidence calibration metrics for BWC-generated signals.
Brier Score
—
ExcellentPlatt Scaling
a=1.00 b=0.00
n=0 (insufficient — identity params)
Graded Signals
0
0W / 0L / 0B · 0.0% WR
Avg Confidence
0.000
vs 0.0% actual win rate
| Confidence Bin | Count | Predicted | Actual Frequency | Bias |
|---|---|---|---|---|
| 0–20% | 0 | 10% | — | — |
| 20–40% | 0 | 30% | — | — |
| 40–60% | 0 | 50% | — | — |
| 60–80% | 0 | 70% | — | — |
| 80–100% | 0 | 90% | — | — |
| Failure Mode | Count | % Total | Avg R | Avg Brier |
|---|---|---|---|---|
| tp1 hit | 0 | — | 0.00R | 0.000 |
| tp2 hit | 0 | — | 0.00R | 0.000 |
| stop loss | 0 | — | 0.00R | 0.000 |
| expired | 0 | — | 0.00R | 0.000 |
| Symbol | Side | Confidence | Calibrated | Outcome | Failure | R | Brier |
|---|---|---|---|---|---|---|---|
| No graded signals yet. Signals are graded after they close. | |||||||
Methodology
Brier score measures the mean squared difference between predicted probability and actual outcome (1 = win, 0 = loss). Scores range from 0 (perfect) to 1 (worst). Platt scaling fits a logistic regression to adjust model confidence toward empirical frequencies. The calibration table compares predicted vs actual win rates within each confidence bucket — a perfectly calibrated model would show actual frequency matching the predicted bin midpoint.
