Last week I wrote briefly about a theory that the Hilton Super Contest serves as a prediction market for NFL games, and I also said I'd track the performance of games that receive over 100 picks. This week, there were 3 such games: Saints -1 (110), Lions -2.5 (103), and Miami +2.5 (102). We'll see how these picks do -- keeping in mind that the Fins pick was one that I almost made. Overall, the top 5 picks from the Hilton contest don't do all that well, although that doesn't disprove my theory, because often the level of support for the 4th and 5th games will fall off dramatically.
But most intriguingly, my Lions pick enjoys by far the biggest margin of support this week (+61, or 103 Lions picks vs. 42 Redskins picks). It's a delicate line to tow, given that I premised my pick in large part on betting against the public. But I'm proceeding under a working hypothesis that Hilton contestants enjoy an advantage against the general public, and that this news is thus good news. At the very least, it means that my thinking this week is consistent with those who actually gamble, for a living (for the most part, at least -- I was inclined to take the Steelers and not at all impressed by the Saints -1 line). We'll see.