Posted on 10/19/2013
It's no secret that I believe math and sports go together. After designing and analyzing my own baseball statistic, I've decided to try handicapping sports games. I'm starting with (American) football because it is hands-down the most popular sport in America and arguably the sport with the most betting action. Depending on the success of this endeavor, I may move on to basketball and/or baseball. Of course this takes a tremendous amount of time, and it's not something I'm jumping into head-first.
I've done quite a bit of research and preparation for this announcement. I've searched for trends in game outcomes relative to closing betting lines. I've analyzed betting strategies to maximize profits while maintaining winning percentages. Home dogs, road favorites, over/unders, and first-five inning bets: I've looked at them all! I've learned nothing.
There are no patterns. Teams go up and down. Vegas handicappers do a great job of noticing trends and adjusting lines to counteract the advantages an ordinary person might come across. If you try to consistently bet a certain scenario, you may have success for a while, but your luck will eventually run out. For example, if you bet on home dogs getting 4 or fewer points in 2008, you could have made a killing. In 2009, you would have lost everything quickly. So far in 2013, you could have easily tripled your money (the season isn't over however). I have convinced myself that in order to be a successful handicapper, you must use trends AND statistics. Unfortunately, getting free NFL teams statistics for the last five years has proven difficult. Therefore, I have a couple strategies I would like to test for the rest of this NFL season. I'm not expecting it to go well. I expect to learn and make adjustment as the season progresses, eliminating poor strategies and coming up with new ones.
I have two separate strategies involving only team statistics used to guess which team should be favored and by how much. There is a third strategy which is an "average" of these in some sense. Over the next ten weeks I will do my best to handicap each game - both the spread and over/under - in hopes of establishing significant statistics which tend to predict the outcomes of football games. I haven't decided if I will post all my picks beforehand, or simply report a winning percentage each week. Either way, the reader should have faith that all of my lines have been calculated before each game and the numbers will never be adjusted to skew in my favor. I really want to know if this is going to work. Since this is the first week, and the first draft of this post was written on Friday night, I will only report results at the conclusion of the Monday night game.
I will also say that I will never sell these picks; I am not trying to become a tout to make money. This is an experiment. If my abilities become fruitful, I will only post my results after the fact for people to see and experiment with on their own. In the more likely scenario that this turns out to be an epic failure, having my ideas displayed will also be very interesting. Well, maybe only interesting to me.
Allow me to introduce myself. I am a mathematician and programmer. Currently, I am working on a PhD in arithmetic geometry. I like to write about many things including math, sports, programming, education, and technology. If you would like to see my comments more frequently, you can follow me on my social profiles: