(Unsuccessful) Handicapping Update
by, 10-05-2011 at 05:00 PM (1068 Views)
Well, my day job is in the investment arena so my blogging has been minimal of date and will probably be sporadic this year. No great loss to the community!
I did want to update a couple of topics, though. This year’s Hilton SuperContest set a record for entrants, assuredly as a result of publicity last year engendered by Bill Simmons entering, writing and tweeting about the contest and Chad Millman not only doing the same but talking about the legal proxy services available for out-of-town handicappers. You have to enter the contest in person but may use a proxy service during the year. Winnings must also be collected in person.
The contest received 517 entries at $1,500 a pop making for a $775,500 total pot, of which 40% goes to the winner. I have entered on behalf of a small consortium and we are off to a slow start. We have gone 2-3, 1.5-3.5, 3-2 and 3-2 and sit at 9.5-10.5 so far, seven out of the lead and four out of the money. It’s early, it remains doable and we are clearly on the comeback trail!
I head up another, larger in number consortium attempting to hit the Ontario Government’s Pro-Picks game where you have to pick the winners of all games, straight-up, each weekend. Last week, a perfect card and a $5 ticket returned $29,444. That kind of hurts, as last week was doable and we whiffed on the Bills and Eagles. I make the call on the ticket and had the Eagles game boxed originally as I am not sold on them, plus I contemplated boxing the Bills. Late changes “improved” the ticket into a loser.
One interesting aspect of attempting to hit both games is using data from the Hilton contest in my Pro-Picks selections. The Hilton contest gets quite a few games each week where the players are all on one side of the boat. For instance, 159 players took the Vikings while only 20 took the Chiefs despite it being a 1-point spread and the Chiefs being home dogs. I used this knowledge to box the Vikes/Chiefs game on Pro-Picks, figuring if the Chiefs came in the ticket would pay much more despite it looking like a pick’em affair that would lead to a 50/50 split. (We’re hunting for big game and like to have winners that few have picked. Generally, that means big dogs but not in this case was my guess.)
I’ve noticed that my Hilton SuperContest team is landing on the steamed games quite frequently and I don’t think that’s a good thing. I think it means we are doing superficial analysis and coming to the obvious conclusions. The Vegas boys are well aware of how the public processes its first-level thinking and everything we are acting on is in the lines already. This is easier to identify in retrospect than to fix going forward but we’re going to give it a try.
I’m a dyed-in-the-wool value investor and am always seeking out undiscovered value and statistical data that is being ignored by the market. As a result, I’m drawn to sites like Football Outsiders and check the DVOA tables every week. I have started to look at Brian Burke’s work at advancednflstats.com and I’ll read any stats-based column. I pay attention to them primarily because they are objective pieces of information whereas so much NFL analysis is just gut-level assertions tied together with pre-existing emotional biases. However, I have slowly come to the conclusion that NFL stats don’t cover enough of the game well enough to be as useful to the NFL analyst as baseball stats are to baseball geeks. On top of that, the NFL is a low-scoring game (21-14 is really 3-2), where single bounces affect outcomes greatly. And, it’s a 16-game season, where individual games affect overall results greatly.
So, my conclusion is that predicting the NFL is a difficult process (shock of shocks) where statistical analysis is helpful but not sufficient. At the spread level, determining the simple public view and figuring out where the public has been overly simplistic or reactive is probably as useful an approach as any. I haven’t found the Football Outsiders stats as useful as I would have hoped in this regard, but that could just be my inferior utilization of them. Or, it could be that they are now being fully incorporated by oddsmakers, rendering them accurate but fruitless. At the straight-up level, detailed personal subjective analysis supplemented by the stats work is my guess as the best approach. Sadly, I ain’t got the time to do that well. Heck, so far, I haven’t really found anyone else who can either. But I’m still hunting and will keep you apprised of any progress I have.