IF you take out CHI (32 points), LAC (29 points), JAC (22 points), LAR (19 points), and DAL (17 points), then it looks like a normal week. Because they were so ridiculously high scoring this week any model NOT having those in the top 5 looks terrible.
Let’s see them:
Model H is apparently my boom/bust model. For consistently good choices it’s still Model F and Yahoo’s pros. I’ll keep using them until I can get one more data point saying I should use Model H.