The record · Horse racing
Horse racing: calibration and closing-line value
Do the probabilities mean what they say, and does the price taken beat the close? Both charts are drawn from the resolved-call ledger by the same rules as the scoreboard, with the sample size beside them.
Reliability diagram
Calls are grouped by the probability they stated. A calibrated forecaster’s dots sit on the diagonal; a bar that misses it is a claim the sample does not yet support. Bins with few calls have wide bars.
aggregateracing.win
n = 5,554 scored calls
Show the bins as a table
| Stated probability | Calls | Mean stated | Came true | 95% interval |
|---|---|---|---|---|
| 0% to 10% | 3,838 | 0.042 | 0.042 | 0.036 to 0.048 |
| 10% to 20% | 1,156 | 0.143 | 0.151 | 0.131 to 0.172 |
| 20% to 30% | 323 | 0.242 | 0.241 | 0.198 to 0.291 |
| 30% to 40% | 123 | 0.340 | 0.398 | 0.316 to 0.487 |
| 40% to 50% | 59 | 0.444 | 0.458 | 0.337 to 0.583 |
| 50% to 60% | 29 | 0.546 | 0.655 | 0.473 to 0.801 |
| 60% to 70% | 18 | 0.640 | 0.889 | 0.672 to 0.969 |
| 70% to 80% | 7 | 0.744 | 0.857 | 0.487 to 0.974 |
| 80% to 90% | 1 | 0.821 | 0.000 | 0.000 to 0.793 |
racing-marketracing.winmarket baseline
n = 5,554 scored calls
Show the bins as a table
| Stated probability | Calls | Mean stated | Came true | 95% interval |
|---|---|---|---|---|
| 0% to 10% | 3,838 | 0.042 | 0.042 | 0.036 to 0.048 |
| 10% to 20% | 1,156 | 0.143 | 0.151 | 0.131 to 0.172 |
| 20% to 30% | 323 | 0.242 | 0.241 | 0.198 to 0.291 |
| 30% to 40% | 123 | 0.340 | 0.398 | 0.316 to 0.487 |
| 40% to 50% | 59 | 0.444 | 0.458 | 0.337 to 0.583 |
| 50% to 60% | 29 | 0.546 | 0.655 | 0.473 to 0.801 |
| 60% to 70% | 18 | 0.640 | 0.889 | 0.672 to 0.969 |
| 70% to 80% | 7 | 0.744 | 0.857 | 0.487 to 0.974 |
| 80% to 90% | 1 | 0.821 | 0.000 | 0.000 to 0.793 |
aggregateracing.win.jumps
n = 2,523 scored calls
Show the bins as a table
| Stated probability | Calls | Mean stated | Came true | 95% interval |
|---|---|---|---|---|
| 0% to 10% | 1,476 | 0.040 | 0.047 | 0.037 to 0.059 |
| 10% to 20% | 570 | 0.145 | 0.151 | 0.124 to 0.183 |
| 20% to 30% | 244 | 0.243 | 0.258 | 0.207 to 0.317 |
| 30% to 40% | 137 | 0.345 | 0.372 | 0.296 to 0.456 |
| 40% to 50% | 52 | 0.448 | 0.327 | 0.215 to 0.462 |
| 50% to 60% | 28 | 0.548 | 0.500 | 0.326 to 0.674 |
| 60% to 70% | 11 | 0.647 | 0.727 | 0.434 to 0.903 |
| 70% to 80% | 4 | 0.754 | 1.000 | 0.510 to 1.000 |
| 80% to 90% | 1 | 0.868 | 1.000 | 0.207 to 1.000 |
racing-marketracing.win.jumpsmarket baseline
n = 2,523 scored calls
Show the bins as a table
| Stated probability | Calls | Mean stated | Came true | 95% interval |
|---|---|---|---|---|
| 0% to 10% | 1,476 | 0.040 | 0.047 | 0.037 to 0.059 |
| 10% to 20% | 570 | 0.145 | 0.151 | 0.124 to 0.183 |
| 20% to 30% | 244 | 0.243 | 0.258 | 0.207 to 0.317 |
| 30% to 40% | 137 | 0.345 | 0.372 | 0.296 to 0.456 |
| 40% to 50% | 52 | 0.448 | 0.327 | 0.215 to 0.462 |
| 50% to 60% | 28 | 0.548 | 0.500 | 0.326 to 0.674 |
| 60% to 70% | 11 | 0.647 | 0.727 | 0.434 to 0.903 |
| 70% to 80% | 4 | 0.754 | 1.000 | 0.510 to 1.000 |
| 80% to 90% | 1 | 0.868 | 1.000 | 0.207 to 1.000 |
Closing-line value over time
Each act compares the price taken with the closing price observed after it. Above zero means the price taken was better than the close. Acts with no later closing price are not plotted.
n = 5 acts · Mean closing-line value +25.9%
Not enough resolved acts yet (n = 5)
Closing-line value needs at least 30 acts with a closing price observed after the decision.
Informational only, not advice and not an invitation to trade or bet. Everything runs on paper. Past results do not predict future ones.