OriginalCanesCanesCanes
All-ACC (#1 most reproted porster on CIS)
- Joined
- Feb 7, 2013
- Messages
- 35,268
I can’t speak to Steele’s models but I know how I do it is by regressing the data towards the mean using past years data while weighting the most recent year more heavily.
I’ll use averages for previous teams and then use generally available data about newcomers to get a starting spot. Then I regress based on recruiting ranking or performance in HS (versus peers) and then complete projections.
Do something similar for data available for coaching staffs.
It’s a difficult process, and the accuracy for individual players is hit-and-miss, but you can get a pretty good gauge of team performance with this method.
Vegas does a similar process in win projection modeling as well.
That’s probably a great way to do it.
I think what people without any statistics background might fail to consider is the limitations of any models with that many variables.
It’s nearly impossible to correlate that many factors, so a ton of assumptions end up getting made.
With that many factors to correlate, it can get as complex and multi-factorial as climate science modeling