With the AFL confirming the Brownlow Medal ceremony is to be held on Sunday, 19th September we dug out our trusted Brownlow Medal machine learning model to see if we could predict the 2021 winner.

Last year we predicted the winner, as well as 4 of the top 5 players. (Read more about it here)

This year one of the potential flaws in our machine learning model may be influencing the predictions.  Our model uses a single player’s game data as an input rather than an entire game’s worth of data.  

Each row contains the data items listed below for one player for one game. That is approximately 15,000 rows of data!

• Frees for less frees against
• Contested possessions
• Contested marks
• Goal assists
• Team winning/losing margin

This year’s predictions show teammates Clayton Oliver (Melbourne) and Christian Petracca (Melbourne) in first and second, although this prediction may not be unreasonable due to the incredible seasons both players have had as well as Melbourne’s success as a team.

Our model also has teammates Marcus Bontempelli (Bulldogs), Tom Liberatore (Bulldogs), and Jack Macrae (Bulldogs) finishing in 5th, 6th, and 7th respectively.  It is less likely that these three players from the same team will be able to manage enough votes collectively to achieve that result.

The Parity Analytic Machine Learning algorithm has used the player data to come up with the following top ten:

Parity Analytic Brownlow Medal Top 10

1.   Clayton Oliver (Melbourne)

2.   Christian Petracca (Melbourne)

3.   Darcy Parish (Essendon)

4.   Luke Parker (Sydney Swans)

5.   Marcus Bontempelli (Bulldogs)

6.   Tom Liberatore (Bulldogs)

7.   Jack Macrae (Bulldogs)

8.   Ollie Wines (Port Adelaide)

9.   Jarryd Lyons (Brisbane)

10. Jack Steele (St Kilda)

Let’s see how we go this year. 

If you are interested in learning more about our Excel based machine learning model you can find a detailed article here.

We are also able to provide a copy of this years’ model, email us at:  brownlow_ml@parityanalytic.com.au