Who's Best To Play At Home?

The 2015 AFL Schedule is imbalanced, as have been all AFL schedules since 1987 when the competition expanded to 14 teams,  by which I mean that not every team plays every other team at home and away during the regular season. As many have written, this is not an ideal situation since it distorts the relative opportunities of teams' playing in Finals. 

As we'll see in this blog, teams will have distinct preferences for how that imbalance is reflected in their draw.

Read More

Predicting the Final Ladder

Discussions about the final finishing order of the 18 AFL teams are popular at the moment. In the past few weeks alone I've had an e-mail request for my latest prediction of the final ordering (which I don't have), a request to make regular updates during the season, a link to my earlier post on the teams' 2015 schedule strength turning up in a thread on the bigfooty site about the whole who-finishes-where debate, and a Twitter conversation about just how difficult it is, probabilistically speaking, to assign the correct ladder position to all 18 teams. 

Read More

Ensemble Encore

The idea of ensemble learning and prediction intrigues me, which, I suppose, is why I've written about it so often here on MoS, for example here in introducing the Really Simple Margin Predictorshere in a more theoretical context, and, much earlier, here about creating an ensemble from different Head-to-Head predictors. The basic concept, which is that a combination of forecasters can outperform any single one of them, seems plausible yet remarkable. By taking nothing more than what we already have - a set of forecasts - we're somehow able to conjure empirical evidence for the cliche that "none of us is better than all of us" (at least some of the time)

Read More

AFL Crowds and Optimal Uncertainty

Fans the world over, the literature shows, like a little uncertainty in their sports. AFL fans are no different, as I recounted in a 2012 blog entitled Do Fans Really Want Close Games? in which I described regressions showing that crowds were larger at games where the level of expected surprisal or 'entropy' was higher.

Read More

On Choosing Strong Classifiers for Predicting Line Betting Results

The themes in this blog have been bouncing around in my thoughts - in virtual and in unpublished blog form - for quite a while now. My formal qualifications are as an Econometrician but many of the models that I find myself using in MoS come from the more recent (though still surprisingly old) Machine Learning (ML) discipline, which I'd characterise as being more concerned with the predictive ability of a model than with its theoretical pedigree. (Breiman wrote a wonderful piece on this topic, entitled Statistical Modelling: The Two Cultures, back in 2005.)

Read More