Sep 16, 2024
11:18:50am
11:18:50am
Or ideally, they could figure out how much to weight preseason priors, using historical data to get the best predictions
Not just arbitrary, but really looking into the historical data across many years and taking a statistical approach to assess how much preseason priors should be weighted as the season goes on, with the algorithm eventually not considering preseason expectations at all once there is enough data.
That really seems like it would be the best case.
That really seems like it would be the best case.
- Bio page
- krindorr
- Joined
- Oct 5, 2020
- Last login
- Sep 19, 2024
- Total posts
- 18,942 (115 FO)
Messages
Time
Sep 16, 7:40am
Sep 16, 10:38am