Round 5 Model Update

Who is the Best Team?

Gather Round! Gather Round! It is time for the AFL's first Gather Round, so let's all Gather Round the model ratings and see who we have on top after four rounds of footy.

Preseason we had Geelong leading Sydney with the former above 1700 elo points and the later slightly behind on 1680. As a reminder, average across the league is and will always be 1500.

The worst teams were North Melbourne on 1197 and West Coast on 1198. These latest updates will be very good news for some of the worst teams and not so good news for the better teams.

As it currently stands every team is above 1300 and no team is greater than the low 1600s. This is partly due to the new season slightly evening out all the teams to account for player changes and unknown varibles in the offseason, but is also down to some of the suprises we have seen this season so far. The AFL have also helped this out with some very even teams being matched up so far through the first four games.

So who is where? Let's find out.

1. Melbourne - 1612.4
2. Collingwood - 1595.7
3. St Kilda - 1586.5
4. Sydney - 1581.9
5. Geelong - 1566.6
6. Brisbane - 1562.4
7. Carlton - 1554.7
8. Richmond 1533.6
9. Western Bulldogs - 1510.3
10. Fremantle - 1500.4
11. Essendon - 1499.4
12. Port Adelaide - 1492.5
13. Adelaide - 1479.4
14. Gold Coast - 1461.7
15. Hawthorn - 1397.93
16. GWS - 1397.88
17. West Coast - 1333.11
18. North Melbourne - 1333.09

Some interesting results to check out here. Firstly Melbourne being top of the pops shouldn't be too much of a suprise, Collingwood second ditto but St Kilda are flying at the moment and complete the podium thanks to their 4-0 start.

Preseason favourites Geelong have started 1-3 and dropped nearly 200 elo points down to 5th place. 

Hawthorn and GWS are an almost identical rating, same with West Coast and North Melbourne, however the 15th and 16th paring are playing off this week at a neutral venue resulting in the closest tip the model has produced to date, a 50.001% probability for Hawthorn to win.

Do Ratings Correlate to Predicted Ladder?

This is the next natural question. It is fine to say Melbourne are first on rating and Geelong are fifth. It means a little bit to suggest Port Adelaide lead their cross town rivals by 13 elo points. But how does this stack up compared to our final ladder prediction as of right now?

Thanks to some team's fixtures being easier or harder compared to others, there is some disparity between the elo ratings and ladder predictor.

The top three are all the same based on their number of expected wins. Carlton move up 7th to 4th, with Geelong dropping from 5th to 8th. Essendon go from 11th into the finals to 7th, while most other teams remain pretty stagnant.

The full ladder with expected number of wins (draws count as 0.5), shown here.

1. Melbourne - 14.6
2. Collingwood - 14.1
3. St Kilda - 14.0
4. Carlton - 13.3
5. Sydney - 12.8
6. Brisbane - 12.6
7. Essendon - 12.5
8. Geelong - 11.9
9. Richmond - 11.5
10. Western Bulldogs - 11.4
11. Port Adelaide - 11.4
12. Adelaide - 11.2
13. Fremantle - 10.3
14. Gold Coast - 10.0
15. GWS - 8.7
16. Hawthorn - 8.7
17. North Melbourne - 8.6
18. West Coast - 7.5

Are Hawthorn going to win 8.7? Probably not. Is any team going to win a fraction of a game that isn't 0.5? Definitely not. The Squiggle run down on why ladder predictions are wrong is very good but I will explain this prediction here.

The chance of this being the actual final ladder from spots 1 to 18 is almost 0%, there is next to no way of predicting it this far out, especially with such a low percentage chance of anyone finishing in that exact spot. 

Even if my model is 100% accurate, which is most definitely is not, no one's is, there is only a 23% chance St Kilda finish top, a 7.5% chance Richmond finish 9th, a 7.4% chance the Doggies finish 10th.

Multiply all these together and you get a 0.0000000000000002% chance of all the positions being accurate (assuming each finishing position is an independent varible of course).

So what is going to happen? We just don't know. We can model and model all night long but at the end of the day we just need to wait and watch and hope our models are as accurate as can be. With that being said, let's bounce the Sherrin for Gather Round and get ready for a good weekend of footy.

Comments

Post a Comment

Popular posts from this blog

Model Performance - Round 3

How Did the Model Perform? Round three threw up quite a few unexpected results, with somewhere between 4 and 5 underdogs getting the victory, depending on which source you use. For the PPP model  it was the best round so far to date in terms of tips and MAE with a -0.11 bit round being second only to round 1's -0.07 bits.  It is too easy to say this year has been unpredictable and not easy to model, the evidence for this however is in the Squiggle  model competition, with only four models in the net positive on bits. Having three straight rounds of negative bits does notr bode well for your model but so far in 2023 we can all be given a pass while the ratings and teams find their running order and settle down going into the first turn and back straight. If the betting markets are your preferred yardstick then we can safely say we have the bookies covered on MAE and bits, only being one behind on tips. For round three alone, the model tipped 5/9, lost 0.11 bits and had a M...
Read more

Model Performance - Round 1

How Did the Model Perform? Throughout 2023 I will be comparing my model to those in the Squiggle model competition with hopes of competing in it next year.  The Squiggle competition is a very well respected group of modellers who all predict the outcome of every match in the AFL season and then analyse their performance using various methods. In this post  I explained how these were but for those in need of a refresher we have bits (confidence based picks), MAE (the average of how many points you were wrong by) and just general tips (how many games did you pick the winner). On Squiggle the top models by bits are The Footycast  who leads the way with 0.79, Elo Predicts  who is second on 0.59 and AFLalytics  a close third on 0.58. My model finished round one in a would be 17th place with -0.07 bits. Not to worry, 23 rounds of AFL football to go. Looking at MAE now and we have  The Footycast   again in first with 27.13, The Cruncher in second with ...
Read more

AFL 2023 Season Preview

Image
2023 Initial Ratings and Model Explaination The 2023 footy season is upon us and it's time for wannabe modellers to break out their coding skills and put their brain and football knowledge to the test. I have refined and improved my previous models for sports like rugby, baseball and tennis to come up with what I believe to be an accurate and realistic model for the AFL season. My model works mainly from elo as its base rating system but includes variables like a fluid home ground advantage multiplier, expected score (or xScore) and previous head to head results to skew the probabilities one way or the other. The Elo Rating System Elo works on a couple main principles. Firstly is that it is a zero sum rating system, which means whatever points you add to one team you must take the same number of points away from the other team.  Secondly there is no "perfect" rating for each team as it is fluid and changes after every game.  For this reason two teams may play 10 matches a...
Read more