Compare AI models
Two models, one window, one asset class, one horizon. Accuracy is shown with a 95% confidence interval, because a rate without one cannot be compared to another rate.
Rule-Based Analysis
- Sample size8410
- Correct1023
- Mean stated confidence 32.0%
- Confidence gap 19.9 pts
- Symbols covered 1020 / 1032
- Coverage of the field 98.8%
- Horizons used1
| Confidence | n | Accuracy |
|---|---|---|
| 0–50% | 8058 | 11.4% |
| 50–70% | 190 | 38.9% |
| 70–85% | 42 | 14.3% |
| 85–100% | 120 | 22.5% |
Mistral Small Latest
- Sample size1071
- Correct390
- Mean stated confidence 74.5%
- Confidence gap 38.1 pts
- Symbols covered 138 / 1032
- Coverage of the field 13.4%
- Horizons used2
| Confidence | n | Accuracy |
|---|---|---|
| 0–50% | 20 | 55.0% |
| 50–70% | 380 | 45.3% |
| 70–85% | 247 | 38.1% |
| 85–100% | 424 | 26.7% |
Over the last 30 days the two confidence intervals do not overlap, so this sample does separate the models.
Accuracy counts a prediction correct when the direction it stated matches the direction the price moved past a 1% threshold over the stated horizon. Coverage is the share of symbols scored in this window that the model expressed an opinion on — a high rate over four symbols is not the same claim as the same rate over ninety. Full method and disclosures