Model Performance - Round 2

How Did the Model Perform?

Round two threw up arguably more inconsistent results than the week previous, with no Squiggle model earning more than 5/9 tips, the mean being just above 4. North Melbourne cut deep into many tipsters bits and hurt a lot of performance metrics, ensuring no one would bank the perfect round, only assisted by brilliant performances by Carlton, Brisbane, St Kilda and West Coast who all won as underdogs.

The PPP model only copped four tips this week, with the best result being a complete fence sit on the Western Bulldogs - St Kilda match up, resulting in no lost bits and a lesser mean than any other Squiggle model. 

Bits

The model lost 0.57 bits this week and had a MAE of 36.22, not a good week by anyone's standards but a "less bad" week than most other models.

On the Squiggle leaderboards we now would sit 8th with -0.64 bits, just above the s10 model, which is a combined model of the previous year's top 10 models. The s10 is more-or-less the benchmark for models so to be above this is reassuring although we have only played 18 games of the 207 regular season matchups.

Leading the way on bits now is a new leader, plusSixOne, who sits atop with 0.36 bits, 0.3 ahead of nearest competitor and last week's leader The Footycast who is on 0.06. The Squiggle model itself is the only other model in the net positive with 0.01 bits.

MAE

Looking to MAE and it paints a worse picture for the PPP model where we currently sit on 32.92 which is good for 18th, down from 12th after week one. The Footycast is again on top here and the only model with a sub 30 MAE. Cheap Stats and AFLalytics, round out the top three.

Tipping and FMI Competition

As discussed we only predicted four games correctly, however there was better news in the FMIplay competition (the promotion/relegation tipping competition) with a 3-1 win over TheAFLBluster, putting as at 1-0-1 in Division 7, good for 7th only on points difference, two spots out of the finals. Good to get the first win and monkey off the back.

We play ZaphBot next week, a fellow model and one who competes in the Squiggle competition. We cover ZaphBot by 1.29 bits and 0.02 MAE while having one less correct tip than it, a deficit of 9 to 10.

Betting

The model went 1/2 on bets this week, winning the 0.5 unit selection on St Kilda. Unfortunately I suggested we take the stake down from 3 units, we have robbed ourselves of a much bigger win than otherwise could have been. The losing bet was on GWS in the final game of the week, a rather disappointing performance from the Tsunami men.

This meant we lost 0.15 units this week, bringing the season total to -2.15.

Wrap Up

Overall it was a bit of a meh week. 4/9 tips, a loss of bits and MAE. We were rewarded compared to other models for not being too aggressive on some tips which scraped back some respect.

Round three kicks off Thursday night with Bulldogs and Brisbane going at it, tips and predictions to come.

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