What predictive power do AFL pre-season results have for the regular season? #NABChallenge

Footy is back!

Sort of!

Pre-season trial games have started at obscure ovals where cricket isn’t still getting in the way. It’s a time of great hope, when every team’s draftees and fringe players and returning injured players are showing good signs, and fans and pundits try to identify which teams might have good years by watching each club’s efforts in practice games.

It is a time when the footy press and public relations machine, grateful to have something other than off-field scandal to cover, does its damnedest to sell hope and hype for the season to come. So we’ve decided to take a quick look at how much of that hope and hype is justified.

We have compared preseason and regular season results, compiling the scores from all official NAB Cup, NAB Challenge and AFL-scheduled practice games (including that hilarious 3-way arrangement they ran for a few years), to see how strong the correlation between February and August looks over the last five years:

psvreg1

psvreg2

Yep, with a simple linear trendline, that’s an R² below 15% for both measures. There’s barely a positive predictive relationship in there at all.

There are so many examples of teams blitzing the preseason and then crashing hard, and plenty of premiership and top 4 sides who did nothing in the pre-season. Pre-season form explains nearly none of regular-season form.

We’re not telling you anything new here, but: how your club performs in terms of results in the preseason tells you basically nothing about how the regular season will go and you’re a fool for ever thinking otherwise.

Leave a Reply

Your email address will not be published. Required fields are marked *