Player Estimated Result Tipping (PERT) – 2019

The Player Estimated Result Tipping (PERT) model predicts games from the selected 22 and how each player has performed in the past, rather than using past results or estimated strengths for the club as a whole as with Elo and related methods. It uses marginal Player Approximate Value (mPAV) to estimate strength in defence, midfield and offence to produce an expected scoreline and margin. Our post introducing the concept explains the method in more detail, and last year’s results are here. We can also be tracked at the excellent Squiggle website along with other modellers. Based on last year’s data, we’ve added a blanket 9 point home game advantage where it’s a genuine home game.

Match
Home team
Away team
Home margin
Home probability
1BrisbaneCollingwood12.363.9%
2North MelbourneEssendon-1.748.1%
3West CoastPort Adelaide-6.242.9%
4GWSFremantle21.472.9%
5MelbourneSt Kilda7.058.1%
6RichmondSydney6.957.9%
7Western BulldogsCarlton31.381.0%
8AdelaideGold Coast1.151.3%
9HawthornGeelong-0.949.0%

Cumulative tips
Percent tips
Mean Absolute Error
1850.0%28.5