2016 AFL Season – The Statscrunch Model

So, a new AFL season is upon us, and it wouldn’t be an AFL season without the intense speculation and predictions made.

Statscrunch is now joining the pack, making its own predictions using a model that can be run (handily) from an Excel spreadsheet.

Before we begin, note that I am not an expert at this. If you do want experts, or at least people with experience, then you should look at Matter of Stats and Footy Maths Institute.

Onto the model. It makes use of two intuitively obvious facts:

  • Teams that score high numbers of points tend to win.
  • Teams that concede low numbers of points tend to win.

Also, teams do better at home than they do away, and momentum can be a powerful thing. These facts will all be used in our simple model.

Our model uses this bit of Excel formula:


B is Atk Pts, C is Def Pts, D is HGA, and E is Strk. So this doesn’t look like a cheap RPG, I’ll explain some things that, frankly, should be quite obvious.

HGA stands for Home Ground Advantage, and teams with it are given a 5% loading. to their score. Strk is Streak, and for every consecutive game you’ve won, you get a 1% loading as well.

Atk and Def Pts are how many points the team scores and concedes per game, on average. Whichever team ends up with the higher score is our predicted winner.

So, for tonight’s Hawthorn v Carlton game, using the 2015 stats:hawthorn v carlton.png

A landslide Hawthorn victory. Stay tuned for more predictions using the model. It will be improved as the season goes on, but please comment with your suggestions!


One comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s