The Brownlow Medal Predictions


Betfair has partnered with a statistician to build a prediction model for the 2015 Brownlow Medal. The statistician now known as ‘The Professor’ has this year’s Brownlow winner likely to be Nat Fyfe, Sam Mitchell or Matt Priddis.

The Professor 

Associate Professor Michael Bailey has worked as a biostatistician for 20 years and has published over 400 manuscripts in peer reviewed medical journals that have been cited more than 14,000 times.

In 2005 he completed a PhD which outlined a statistical approach to predicting sporting outcomes and with a specific chapter dedicated to the Brownlow, he is the only known person to be awarded a PhD for the prediction of the Brownlow Medal winner.

Prior to 2015, Fremantle midfielder Nat Fyfe had polled 75 Brownlow votes from only 83 eligible matches.

Average of 0.90 votes per game, there are only 2 players in the entire history of the game that have a higher career average of votes per game (Haydon Bunton & Graeme Moss).

2015 Results: Ranked by the model as being in the top 3 players on the field in 15 matches, including a remarkable run of the first 13 matches in a row.

With 8 potential best on ground performances, he is the player most likely to win the 2015 Brownlow medal.

Winner Market: $2.06

2015 Brownlow Medal Predicted Top 6


Brownlow Medal 2015 Market 

The Hawthorn great is Ranked 9th on the all-time list of most Brownlow votes ever polled with 178 career Brownlow career votes. He’s most likely going to move into the top 5 most votes of all time after this year’s count. The only current player to poll more votes is Gary Ablett.

2015 Results: Ranked in the top 3 players on the ground on 14 occasions

Whether he can garner enough votes to defeat Fyfe, may well hinge on the outcome of the Round 15 clash between Hawthorn and Fremantle. With 39 possessions Mitchell was ranked as the leading player on the ground.

It was an errant knee by Mitchell to the thigh of Nat Fyfe that was arguably the catalyst for a string of injuries that would go on to hampered Fyfe for the remainder of the season and may potentially cost him a Brownlow. Watch out for an awkward podium question should Mitchell get over the line.

Winner Market: $12.5

Won the 2014 Brownlow after being ranked by the model as being in the top 3 players on 11 occasions.

2015 Results: Ranked in the top 3 players on 14 occasions so has clearly had a better season.

If successful, he will be the first player to win back to back Brownlow medals since St Kilda’s Robert Harvey (1997 & 1998).

Winner Market: $7.00

Other Contenders 

Sydney midfielders Dan Hannebery and Josh Kennedy round out the top 5 predicted players for the season.

With both players being ranked in the top 5 players on the ground on 14 occasions, and the top 3, 11 times, they are difficult to split as the leading Sydney vote catcher and either could win the 2015 Brownlow if luck goes their way.

With a career average of 0.60 Brownlow votes per game, Adelaide’s Patrick Dangerfield ranks amongst the top 10 current AFL players for votes polled per game. However, it is his ability to poll 3-votes in comparison to 2 and 1 votes that stands him above his peers.

Winner Market: Patrick Dangerfield $11.50

With 18 3-votes, 8 2-votes and only 5 1-votes, his is amongst only a handful of player to have ever polled more 3-votes than 2 and 1 votes combined. Having polled more than 20 votes in each of the past 3 seasons, and having his best season to date, he must also be given a realistic chance of success.

Predicted Top 12 Players for 2015

Top 12 Players 2015
Ranked Top 5 = Number of times each player was ranked in the top 5 players on the ground
Ranked Top 3 = Number of times each player was ranked in the top 3 players on the ground
Ranked Top 2 = Number of times each player was ranked in the top 2 players on the ground
Ranked 1 = Number of times each player was ranked as the top player on the ground

Betting Activity

Value for Sam Mitchell – predicted a $4.00 chance yet $13.00 in the current market odds. Big value for Mitchell at the win odds or take a safer approach with the top 3 or top 5 markets.

Patrick Dangerfield too short – predicted a $20 shot  through the Model yet sitting around the $10 mark currently. Good quality Lay in the Winner Market.


Brownlow Winner Market

Brownlow Top 3 Market

Brownlow Top 5 Market

How the count will eventuate

With Nat Fyfe expected to poll votes in the first 13 matches of the season, he is likely to be clearly ahead in the count by round 4 and remain the leader for the duration of the count.

If he does get beaten for the 2015 Brownlow, it is unlikely to occur until Round 23 where Mitchell, Priddis, Hannebery and Kennedy are all likely to poll well.

Brownlow Graph

Predicted Top Votes by Team

Historically, when considering individual teams, the leading vote getter for each team was ranked first by the model 84% of the time, with the 2nd ranked player by the model getting the most votes 11% of the time.

Top by Team
Ranked Top 5 = Number of times each player was ranked in the top 5 players on the ground
Ranked Top 3 = Number of times each player was ranked in the top 3 players on the ground
Ranked 1 = Number of times each player was ranked as the top player on the ground

Method in Predicting the Brownlow Winner


The Brownlow medal has been held annually with the 2015 Brownlow being the 88th count. There have been 100 previous Brownlow medals awarded to 82 unique players.

There have been 11 instances when the Brownlow was tied (dual winners 9 times, 3 winners twice) and there have been 14 multiple winners, the most recent of which being Gary Ablett (2009 & 2013), Chris Judd (2004 & 2010) and Adam Goodes (2003 & 2006).


In 1993 the AFL introduced a third field umpire. Since this time, there has been a small but statistically significant increase in the number of votes polled by the Brownlow winner, with the winning total increasing on average by about 1 vote every 3 years (0.32±0.12 per year p=0.02).

Coincidently, all Brownlow medals over the past 20 years have been won by midfield players who typically collect a high number of possessions throughout the course of the match.

Winning Vote Total


By using past information, it is possible to identify specific match and player features that are related to who will be awarded votes.

For example, 90% of all 3-votes, 78% of all 2-votes and 74% of all 1-votes are awarded to players from the winning team, with the probabilities further increasing with the margin of victory. Similarly, the number of possession that each player gets is strongly related to the number of votes polled, with the leading possession winner for each match receiving 3-votes 40% of the time.

The natural ordering of the voting structure (3-2-1-0) lends itself to the use of a mathematical modelling process known as ordinal logistic regression. With a wealth of player information now readily available, by using information from over 2000 past matches, it is possible to identify and combine the effects of more than 20 contributing variables to produce a probability that each player will poll 3, 2 or 1 vote in each match.

The main contributing variables in the model include features of the game such as kicks, marks, handballs, goals and margin of victory, however features relating to the specific player such as quality of possession and past voting performance are also important contributors.

By combining these probabilities, each player is assigned a score from 0 to 3 for each match. By aggregating these scores over the entire season, it is then possible to assign each player with a predicted total of Brownlow votes for the season.

How well do the models perform?

Per game level

The model is remarkably good at identifying the leading players in each match, with the top 5 ranked players accounting for 96% of all 3-votes, 85% of all 2-votes and 72% of all 1-votes.

Furthermore, the leading ranked player in each match as determined by the modelling process will successfully poll 3 votes 55% of the time, 2 votes 23% of the time and 1 vote 10% of the time.

When matches are further categorised according to whether there was 1 clear standout player (as opposed to 2 or 3 standout players) the leading ranked player will successful poll 3 votes in 66% of all games.

Per Season Level

When individual match scores are aggregated for the season, the modelling process can accurately predict the season tally of votes for 84% of all 666 players to within 1 vote.

However, it is important to note that more than 70% of all footballers each season will not poll any votes at all. When considering the top 10 leading players each season, the modelling process will typically identify about 7 or 8 of the top 10 players and 3 or 4 of the top 5.

When considering the winner, over the past 13 season, all winners have come from the top 4 ranked players each year, with the leading ranked player winning on 5 occasions, the second ranked player winning on 5 occasions and the third ranked player winning twice.

Top 10 Players for the  past 5 Years Actual and Predicted Order