So whilst you are all busy making your last minute adjustments to your SC/DT teams, here at TipBetPro we are getting ready to use SC stats (amongst others) to predict the outcome of every AFL match. We’ll be sharing these tipping predictions every week on our website (http://www.tipbetpro.com/) along with key statistical insights right here at TooSerious.

Last year we tipped 152 games correctly (72%), which was 4 more than the bookies favourites, 8 more than the Swinburne University Super Computer (we use fairly standard laptops!) and we finished 1,665th out of nearly half a million footytips.com.au users. Furthermore, 2015 was our worst year for a while – previously we have managed to tip around 75% correctly and TipBetPro finished in the top 200 on footytips.com.au in both 2013 and 2014.

What is the TipBetPro approach and why does it work?
Most AFL models will only use "team" based factors in modelling a predicted result, i.e. ladder position, form, home ground advantage etc. TipBetPro takes into consideration not just the standard team factors but also the individual characteristics of every player named to take to the field – including SC stats, player ranking and games played.

So unlike other punters, we put a heavier focus on the main factor which determines the result … who is named in the 22 that week! Our experience across other sports has shown this is consistently underestimated by most “models” and that includes those used by the bookies (we also use the model to find the best value bets each week).

It sounds obvious, and in commentary on every forum, you’ll see supporters predicting outcomes and linking it to the latest injuries or who’s returned from suspension. To a certain degree others do factor these in – however we have found that typically the impact has been understated or not enough consideration has been taken of the depth players coming in to replace those out of the side.

Our ladder predictions – based on 5,000 simulations:


We have run 5,000 simulations of the 2016 Home and Away season with a random injury generator to test the depth of each team and determine the expected wins and probability of making the top 8. These ladder predictions not only take into account TipBetPro’s current player rankings of each team’s best 22, but also projected player improvements and declines based on each player’s age and experience.

Without going through every team, our biggest predicted upset when compared to the bookies market (or the predictions from fellow TooSerious members in the predictions thread) is that we consider it unlikely that Sydney will make the top 8 in 2016. Sydney are a fascinating case study and there is no doubt they have remarkable elite talent, notably that they are the only side to match the Hawks with five Top 40 ranked players - Lance Franklin, Josh Kennedy, Jarrad McVeigh, Kieren Jack and Dan Hannebery, not to mention Luke Parker rapidly approaching that group. However, Sydney's depth has taken a hammering in recent seasons and the trend will continue in 2016 with the loss of Lewis Jetta, Adam Goodes, Mike Pyke, Rhys Shaw and Craig Bird. The Swans have a large group of below average starting 22 players and their youth have thus far under-performed compared to their cohort (other than the supremely talented Isaac Heeney and to a lesser extent Jake Lloyd). All this leads TipBetPro to anticipate that Sydney will be the big sliders in 2016 and only 14.1% of our simulated seasons have the Swans making the Top 8 (only 0.6% for the Top 4).

Thanks again to TooSerious for letting us share our predictions, and as always, we welcome all feedback or questions on these predictions or anything else you find on our website – and for those with an interest, we are also doing weekly NRL tips based on a similar approach.

Cheers
Kim

http://www.tipbetpro.com/ - your statistically tested guide to tipping and betting on AFL and NRL