Tracking Tradesports odds

Some thoughts on the Players Championship at the end of day one. I followed the tournament today on my computer, watching both the leaderboard and a website called Tradesports (plus, for about an hour, the television). Tradesports is a website that allows you to bet on sports, politics, and other events, and, as its prices are the product of people putting their money on the line by trading contracts with each other (basically, placing bets), it provides a pretty accurate snapshot of what the odds are of certain events happening. (It’s very helpful in figuring out how strong various presidential candidates are, for example.)

Before the beginning of the Players Championship, the market had Tiger’s odds of winning around 26% and Mickelson’s around 6%. (These odds are implied from the fact that a contract that would pay 100 if Tiger won cost 26 and a contract that would pay 100 if Phil won cost 6.) I thought 26% was an accurate estimate of Tiger’s chance of winning, even though I didn’t feel Tiger was going to win this week. (I know, it’s easy to “remember” predictions like this after Tiger shoots 75 in round 1.) Maybe that feeling came from a combination of Tiger’s mediocre record here, the windy forecast, the huge field, and the fact that Tiger didn’t seem to have his “A” game in last week’s win. I thought Phil’s 6% odds were about right too (maybe a percent or two low?), realistically balancing his strong finishes in the past two weeks against his horrible track record on the TPC’s 17^th hole and his difficulty closing the deal since Winged Foot.

What surprised me as the day went along was that, as accurate as the odds were at the outset (in my opinion), I didn’t think the odds adjusted properly as new information came in. With Tiger especially, there seemed to be some “stickiness” in the price of his contract. When Tiger was +1 through 14 holes, he was still at nearly 20% odds to win on Tradesports. Meanwhile, Mickelson was at 10% when he was -2 through 10 holes. The market was saying that Tiger can spot Phil 3 shots over 3 ½ rounds and Tiger is still twice as likely to win. This sounds even more wrong when you factor in that Phil was clearly playing well and Tiger not so well. It’s true that you can never count Tiger out, but the fact is that when he starts off not playing well, his odds of winning fall sharply – and probably more sharply than 26% to 20%. I think Tradesports generally provides good information, but sometimes it's a little off. It’s fun to try to figure out when those times are.