Is it time to modernise the AFL Draft?

With today’s AFL Draft operating  under very similar guidelines as the original drafts that began in the United States in the early 1900s, (VU) PhD student Jemuel Chandrakumaran is investigating ways to improve the draft that will benefit both players and clubs.

Extending the contracts from two to three years

Mr Chandrakumaran’s research, published in peer-reviewed journal PLoS One, suggests giving AFL clubs the option to add a further season to the draftee contract at a predetermined compensation package should the club choose to do so at the end of the intitial contract.

“This allows teams to have the option of keeping a draftee longer than the initial two years but it doesn’t impede further on a player’s ability to move,” Mr Chandrakumaran said.

The AFL announced in September the next collective bargaining agreement would include three year contracts for draftees, however the VU proposal is about providing choice rather than mandating it.

As the research shows, moving between clubs doesn’t always result in the best player performance.

“Nearly 80 per cent of players tend to stick with their clubs, and generally that works best for the club and the player.

“We were able to model whether moving to another team after two years was actually good for the player – we found it was, but only if you were in the first round of picks one to 16. In that scenario, the player will  get more game time and more money.”

“Anyone else it’s best to stay with the team that initially drafted you,” he explained.

A new way of valuing AFL draft picks

The AFL Draft Value Index (DVI) was introduced in 2015 – the index creates a numerical value to each individual player draft pick determined by historic player compensation or wage and salary data.

Mr Chandrakumaran with researchers from RMIT and Athletics Australia has presented an alternative to the DVI which they believe is fairer to the players, also published in PLoS One.

“Rather than using career compensation as the determinant of value, we use measures of player performance. We developed models to predict on-field performance, such as games played (both in a career and season) after a draftee was selected for the first time by a team,” Mr Chandrakumaran explained.

Using a panel data set, they retrofillted expected performance against draft picks,  controlling for a variety of factors including selection bias.

“We found player salaries did not always strongly correlate to performance and the change in performance between players selected at different points in the draft did not vary as much as their wages.”

“So it’s good for clubs because they don’t need to option all of their picks for one player, and it’s good for the player because they don’t have this unreasonable expectation to perform – we have seen what that kind of pressure can do to an 18 year old kid,” he said.

While this model was applied to the AFL experience, it could be translated to other player drafts.