Typically I want my models to do as well as possible. That doesn’t mean they should be right every time – if I pick a team because they should win 80% of the time, I want to be wrong 20% of the time. Given that I have to be wrong, I’m happier if the models pick the Lions to lose but they win, which is what happened this weekend. And in terms of getting it right, I’m pleased that the models came down on the right side of the Broncos-Jaguars betpocalypse. Other than that, I’ve managed to watch about two hours of football each of the last two weekends (including Thursday and Monday games; this week I saw some of Bears-Giants) thanks to weddings and baseball playoffs. So I’ll keep things pretty tight to how the models did and the updated rankings/season predictions.
I’ll start with the Hilton SuperContest picks. This past week Luigi had the Bills, Titans, Cardinals, Rams, and Jaguars for a 4-1 record. Arizona missed a push with San Fran by a point, but you can’t complain about 4-1. Since I usually follow Luigi, that would put my SuperContest ‘entry’ at 18-11-1 for 18.5 points. After a few weeks of slowly moving up the chart, 18.5 puts me in a tie for 30th, which is the money cut-off. First place has 21 points, so not a big difference, although you can obviously only make up so much ground at a time. Hopefully these top-five picks keep doing well. Yoshi 1 went 3-2 (didn’t get the Rams and instead missed on Cleveland) for a total of 15-14-1, and Yoshi 2 went 4-1 (same picks as Luigi) for a total of 17-12-1.
How about the predictions as a whole? In all games, Luigi was 9-6, but would have passed on two games and gone 7-6 on the rest. That gives Luigi a total of 42-33-1/37-28-1. Yoshi 1 was 7-8/7-7 for a total of 38-37-1/33-29-1, and Yoshi 2 was 8-7/6-5 for a total of 38-37-1/32-28-1. The consensus picks, where all three models agree, were 6-3 for a 26-17-1 total.
Using his lines for the past weekend, Bill went 6-9. Luigi and Yoshi 2 went 9-6 and Yoshi 1 went 8-7. That makes the counts Bill 36-36-4, Luigi 41-31-4, Yoshi 1 41-31-4, and Yoshi 2 39-33-4. So all three models are ahead of Bill, who is treading water at dead even.
Luigi was 4-10 on moneyline bets for his first losing week, and it was a doozy. Usually hitting a few upsets keeps things close even if the actual number of hits is low, but it so happens this week that three of the hits were on favorites. The only underdog hit for Luigi was the Rams. Yoshi 1 and Yoshi 2 were also pretty brutal.
How about some rankings?
Not necessarily a lot of movement at the top, where scores are relatively tight outside of the Broncos. Your non-Denver best offenses are the Eagles, Cowboys, Packers, Bears, and Saints before a little bit of a drop-off. On defense the top five are the Chiefs, Seahawks, Panthers, Patriots (somewhat surprisingly), and Bengals/Colts in a rough tie. From my rough following of games, I would not have necessarily picked the Panthers or Pats. But they’re 4th and 6th on football-reference’s defensive expected points added, so at least that’s some agreement. Both teams are doing well in terms of points surrendered, both are above-average in yards per play against, and are above average on net yards per pass, which takes sacks into account. So I guess there’s something to it.
The Broncos should win another 7 games, and with the 6 in the bag they’re our presumptive best-record team (note that the expected win column is if they played the whole season again). The Seahawks, Saints, and Chiefs aren’t far behind at over 11 though. I still have the NFC North looking pretty competitive with the Packers and Lions aimed at 10 wins and the Bears looking at 9. But it’s a little early to be projecting these numbers with much confidence.
How about important games for the weekend? Cardinals-Seahawks is important because the game is in Arizona, and a Cards win would put them pretty solidly in the NFC West chase. Atlanta-Tampa has some morbid curiosity factor going for it, as there aren’t too many times you have a one-win team as almost a touchdown favorite (as Luigi sees it). Jets-Pats should have some of its usual appeal, and we’ll have to see if the game is as ugly as the first one was. Eagles-Cowboys will be important in deciding who wins the division, as the Giants are out of it and the Redskins don’t look to do much better. I’m going to have an eye on Panthers-Rams just because the model projects Carolina to be very frisky in the remainder of the season. Finally, Colts-Broncos should have a number of storylines going for it, and with the game in Indianapolis the Colts might have something of a shot at winning.