A few of our factors weren’t contributing at all. They impacted our results more than we expected. V3 fixes that — same model, more of it actually working.
We also found two factors that were working backwards. Corrected those. And gave the model a little more room to breathe before we start grading it.
| Season | Week | W/L | Accuracy | Cumulative |
|---|---|---|---|---|
| 1999 | 1 | 7/15 | 47% | 46.67% |
| 1999 | 2 | 8/15 | 53% | 50.00% |
| 1999 | 3 | 9/14 | 64% | 54.55% |
| 1999 | 4 | 7/14 | 50% | 53.45% |
For Science and the love of the game!!! What you do with this data is up to you!!!