I created a new model for NBA post allstar break. It only uses a little over 200 "fresh cases" without the addition of the previous 458 games. After checking this in a limited sample, it appears to predict sides better than totals. It's a young model; we'll see how it does.
2.23.20
LA-6.5 (it's now 7, I'd take it at -7)
Tor-5.5
Minn+13.5
Det +5.5
(I'm waiting on GS to get to +10. If they get to +10 then I'll try to add them. If not, then I'm on the fence about playing them.)
2.23.20
LA-6.5 (it's now 7, I'd take it at -7)
Tor-5.5
Minn+13.5
Det +5.5
(I'm waiting on GS to get to +10. If they get to +10 then I'll try to add them. If not, then I'm on the fence about playing them.)