Quote of the Day: Gabriel Gonzaga - "I'm Not Afraid of Trading Strikes"

In gambling, the odds on display do not represent the true chances that the event will occur, but are the amounts that the bookmaker will pay out on winning bets. In formulating his odds to display the bookmaker will have included a profit margin which effectively means that the payout to a successful bettor is less than that represented by the true chance of the event occurring.

Any isolated moneyline may not represent the objectively “true” odds (which are unknowable). But markets – and risk investments involving fluctuating prices are markets – often have an uncanny ability to converge to values that approximate “true” values (in this cae, win probabilities). Even though individual people have little information, collectively, they have access to more information than any “expert.” This is why in every sport, win probabilities implied by major betting lines always outperform expert predictions. There is an entire scientific literature devoted to this in the field of judgment and decision making. Individual predictions are subject to all sorts of cognitive biases. When a moneyline says a win probability is 70%, the chosen fighter usually wins close to 70% of the time. I did a little comparison awhile back with mma-elo, probably the best rating system in mma (and in many other sports). It’s based on an underlying probability model , so for any given matchup, you can get a predicted win probability that can be directly compared with those implied by moneylines. As good as ELO is, it still showed substantial “miscalibration” for mma (an important shortcoming, considering that the rating differentials that directly imply win probabilities are at the core of the ELO system) – it showed substantial underconfidence in predicting winners – i.e., when elo said the win probability was .9, it was actually more like .8, and the miscalibration was more extreme for more lopsided matches. The betting line probabilities, in contrast, showed excellent calibration (as determined by a variety of statistical techniques developed for evaluating weather forecasts, including Brier score analysis and Murphy decomposition of the Brier score).

In sum, believe the moneylines. The implied winner at least. We can debate all day about whether the actual probability is 65% or 68% (which is not what the Team Link coach is doing), but in the long run, the only people who will be able to truly capitalize on such a small amount of error in predicted values are true experts of the “experts”. Gonzaga is the underdog here.

wut.

Werdum, Struve and Yvel > Wain, Wellisch and Gonzaga

Also, Carwin pisses people off with his holier-than-thou attitude, so some of the undervaluing may not even be related to the fighting itself.