Obviously I should stop talking about specific games. In the past few weeks I’ve put a little more detail into Atlanta- Tampa (models roughly agreed with the spread and Atlanta won in a huge blowout), Giants-Washington (models liked Washington; Giants won handily while I sat Larry Donnell), Vikings-Packers (models like Minnesota; Packers looked like the Packers are supposed to), and Bengals-Patriots (models like Cincy; the line even swung to favor Cincy; the Bengals got thrashed). So if the article for Thursday’s game just has some power rankings and an ‘enjoy the game!’ tagged on the end, now you know why. Let’s see how the models did on every game I didn’t reason through.
Starting with the Bills again (Barnwell currently filling in for Simmons), I again gave him the Thursday night freebie. Barnwell came down to earth a bit, going 8-7 this week (compared to last week’s 8-4-1), but still putting the Bills at a solid 41-34-1. Luigi did one better, getting Carolina, Cleveland, St. Louis, Tampa, Houston, Buffalo, Pittsburgh, KC, and San Diego for a 9-6 record and 39-36-1 total. Yoshi 1 got the Giants and Colts instead of Rams and Steelers, but also went 9-6 for a 34-41-1 total. Yoshi 2 had Luigi’s picks except for Pittsburgh, so went 8-7 for a 35-40-1 total.
In my usual spread picks, Luigi was 5-6-1 (the SBR line was 49ers -5 for the push). Washington was one timeout or 30 seconds away from a second push, but alas. Luigi’s total record is 26-23-3. Yoshi 1 went 6-5-1 for the ever-so-slightly-positive week and an overall record of 21-26-3. Yoshi 2 picked an extra game and got it wrong for a 5-7-1 week and an overall record of 23-26-3. Not a great start for the models this year. The consensus picks again didn’t do great, at 3-4-1 for a total of 14-17-3.
Again, as you might imagine with the mediocre record so far, the SuperContest picks didn’t do well. All three models went 1-4. This is very unusual; the games with the biggest differences typically do better than games with predictions closer to the spreads. This season so far, it’s been the opposite. Luigi’s SuperContest record is 9-16, which compared to his record above of closer to 26-23 means that the ‘iffy’ games have gone something like 17-7. It’s all backwards. Yoshi 1 is 12-13 and Yoshi 2 is 11-14. I imagine these have to turn around soon. Like maybe this week?