Hi Evan, I really enjoyed this article and love the concept.
I do something similar on the CFB side, with a process that tracks ~ 60 different projection models with a lot of the same KPIs. Congrats on such a solid MAE in your model, that's not easy to do.
always thought EvanMiya was just a cool name but it was actually a person woah
Hi Evan, I really enjoyed this article and love the concept.
I do something similar on the CFB side, with a process that tracks ~ 60 different projection models with a lot of the same KPIs. Congrats on such a solid MAE in your model, that's not easy to do.
https://public.tableau.com/app/profile/andrew.percival/viz/CFBPicker/Standings
I agree on the ensemble approach, that's what I do in college football, and the results are similar in that it outperforms most of the single inputs.
https://public.tableau.com/app/profile/andrew.percival/viz/CFBMetricsConsensus/All
Was the ensemble model for game predictions fully out-of-sample?
This reports in-sample, but I just added a footnote that shows the out-of-sample results too, which are very similar.
I'd be curious about how the three individual models and various blends of the three performed relative to vegas spreads and final results.
Bringing in Vegas is worth a whole separate article which I'll have to save for another time.