Having already covered the real life losers from the weekend, let’s move on to the pretend winners and losers: me, Bill Simmons, and the models. I’ll start with the Hilton SuperContest. For the most part, it’s been a rough start. The top three consensus picks this week, for example, were the Packers, Vikings, and Giants. They all lost outright, let alone covered. The number one entry has 12.5 points on a 12-2-1 record, because you have to figure someone would be doing ok, but it drops quickly to 7 people with 11 points, 4 with 10.5, and then 12 with 10. Throw in a few 9.5s, and you hit the top 30 that would be in the money if the season were over.
I used a different model to pick week 1 games and went 2-3; the three models each went 4-1 in week 2. This past week Luigi went a shoddy 1-3-1, with the Colts covering (actually winning outright in a stunner) and the Chargers pushing but the Texans, Bucs, and Raiders coming up short. I was optimistic about the Raiders, but they ended up losing by 16, just off the spread of 15. Since I usually follow Luigi, that would put my SuperContest ‘entry’ at 7-7-1 for 7.5 points. Yoshi 1 also went 1-3-1 with the same winner and push but a couple different losers, and Yoshi 2 went 1-3-1 with the same picks as Luigi. So not a great start to the season, but it’s been rough all over. I have confidence that things will pick up.
How about the predictions as a whole? In all games, Luigi was 9-6-1, but would have passed on two games and gone 7-6-1 on the rest (the Lions and Jets both covered their close games). That gives Luigi a two week total of 17-14-1/15-12-1. Yoshi 1 was 7-8-1/5-6-1 for a total of 16-15-1/14-12-1, and Yoshi 2 was 7-8-1/5-6-1 for a total of 14-17-1/12-13-1. The consensus picks, where all three models agree, are 9-9-1 through two weeks of picks.
Bill Simmons was 14-17-2 coming into this week, but that includes his week 1 picks that I don’t have a good analog for. So Bill is 7-8-1 on my count. Using his lines for the past weekend, Bill went 6-8-2 (the Raiders pushed on his number, along with the Chargers). Luigi and Yoshi 1 went 8-6-2 and Yoshi 2 went 7-7-2. That makes the two week counts Bill 13-16-3, Luigi and Yoshi 1 17-12-3, and Yoshi 2 15-14-1. So we’re off to a solid start.
Luigi was 7-9 on moneyline bets for a positive week, buoyed by the Colts and Browns wins. The Yoshis were in the black as well, but only enough to put Yoshi 1 a bit in the black overall and Yoshi 2 at just under even through two weeks.
Ok, so now that we’re through that, how about the league? Everyone has played three games, which isn’t a lot, but let’s go ahead and roll out the power rankings. Remember that the numbers themselves are somewhat meaningless (they come from Luigi’s regression weights), but by the same token they are tied directly to the predictions. I also include an ‘average’ team in the rating, but you can see that most teams are above-average. That’s because the average team is from the entire data set, and my guess is that the increase in both pass attempts and pass efficiency over the past few years makes recent teams look better.
So the best team in the league right now, as far as Luigi is concerned, is Denver. That seems about right. Next up are Seattle and, perhaps more surprisingly, the Packers and a few other teams that aren’t 3-0. Maybe the most surprising rating is the Chiefs, who are down at 7th. Why? Well, their defense is doing great, which meshes with the general perspective. Their offense has been good but not great, which leads to their ranking. The Chiefs haven’t needed it to be so far because of the turnovers they’ve taken advantage of, but those tend not to last forever. At the other end of the list, we have sad sacks Jacksonville, NY Giants, and Tampa Bay.
While Denver is (theoretically) the best team in the league, does that mean they’ll end up with the best record? Maybe not.
That list has the games actually won by each team so far, how many Luigi expects them to win total (actual plus predicted future wins), how many Luigi thinks they should have won so far, and that predicted future win number. According to Luigi, Seattle might end up 13-3 while Denver is only 11-5 or 10-6, not too different from the Saints or Colts. Why? It’s because while Denver may be a favorite the rest of the way, that doesn’t mean they’ll actually win every game. Peyton will throw an interception at some point, or someone will fumble, or the opponent will get a kick return touchdown. All of those might happen. They might just play some tough teams; looking ahead, Luigi thinks that Denver might have close calls with a number of their road games (the Colts, Pats, Chiefs, etc). In contrast, Seattle is a good team that gets to play an easier slate. These numbers will change each week as we get a better sense not only of how good the Broncos and Seahawks really are, but also their opponents and the quality of the games that have already happened. But what I’m going to take away from the season predictions right now is that the Lions might go 10-6, and that would be great.