Already a subscriber?
Not a subscriber yet?
Man versus Model
Win Probability
Compare LaxElo win probabilities against the probabilities generated by the wisdom of the crowd. LaxElo is an Elo-style algorithm that produces a team-strength rating. When two teams meet, their LaxElo ratings can be converted to a win probability for both teams.
Total Goals
The pace/efficiency model uses my opponent-adjusted efficiencies, combined with the pacing/possessions history for both teams, to estimate how many goals are likely to be scored in a given game. Beware though: it does not factor in weather conditions.
One of the first things I did when I started LacrosseReference was to implement an Elo model to rank teams. For years, I've used the Elo model to calculate win probabilities for any matchup. I work hard to calibrate the model so that an 80% chance means that you should expect 8 wins and 2 losses.
I don't personally bet on games, but one of the big benefits of the rise of lacrosse gambling is that there is a numeric representation of the public's perception of almost every game (at least for DI Men). Since the models are, by definition, calibrated to actual outcomes, the next frontier is seeing how they compare against what the public thinks about games. And the GameGuide was born.