AFL 2010 Season - Brownlow Medal Predictions

Here are my Brownlow predictions for this year. Bit of a strange one this year as I more or less only rate 2 players any chance to win it - Swan and Ablett.

This year I've got 50 individual player stats at my disposal, for every player in every game. Obviously not all of these stats will come into the model but it is a big increase in previous years where I had maybe a dozen stats.

If you download the link below it should open a spreadsheet with the usual Brownlow betting options available - overall winners, number of winners, winning numbers of votes, top votegetter in the team, and various head to head and group betting options. With a lot more effort then it would seem it takes, I've been able to provide - for every player in every game - their chance of polling 3, 2 or 1 votes - which obviously leads to their expected votes per game, and also the chance that player receives any votes. With this you can narrow down to certain players, and on the night track to see if they are ahead or behind expectations - potentially useful for those who want to bet in-play.

Like I say, never before has my model come up with only 2 possibilities, though I've looked game by game and seem pretty happy with what the model is coming up with. I'm pretty hesitant to release these because of this situation, but my model has been pretty good in the past - as always though I can't stress highly enough that these predictions should at best be used as a guide, and certainly not the be all and end all in terms of placing your own Brownlow bets.

So how has the model gone in the past? Pretty good if I say so myself. Last year I rated Ablett a $1.33 chance - much shorter than the $3 available on the day. There's also a lot of value to be found in the exotic options - for example last year I rated Winderlich a $5.80 chance to win Essendon's count, big value seeing as I was able to snap up $34 on the day - even though he ended up sharing the honours with Jobe Watson, it was still a great result.

Anyway have a browse and if there's any queries please let me know. Also if there are any other betting options you'd like to see let me know too and I'll do my best to update the sheet.

2010 Brownlow Medal Predictions

Player	Exp Votes	% Chance
Dane Swan - CW	32.5	50.2%
Gary Ablett - GE	32.3	49.5%
Luke Hodge - HW	19.0	Daylight
Leigh Montagna - SK	18.3	Daylight
Matthew Boyd - WB	17.3	Daylight
Chris Judd - CA	16.2	Daylight
Brent Harvey - KA	15.7	Daylight
Brendon Goddard - SK	15.2	Daylight
Paul Chapman - GE	13.9	Daylight
Michael Barlow - FR	13.6	Daylight
Jonathan Brown - BL	13.2	Daylight
Scott Thompson - AD	12.4	Daylight
Joel Selwood - GE	11.9	Daylight
Aaron Sandilands - FR	11.3	Daylight
Matthew Pavlich - FR	11.2	Daylight
Adam Cooney - WB	11.0	Daylight
Adam Goodes - SY	10.9	Daylight
Lance Franklin - HW	10.8	Daylight
Scott Pendlebury - CW	10.6	Daylight
Nick Dal Santo - SK	10.6	Daylight
Andrew Swallow - KA	10.2	Daylight
Kieren Jack - SY	10.2	Daylight
Jobe Watson - ES	10.0	Daylight

AFL 2010 Season - Round 13 Predictions

A great week last week for the model.  It managed to tip 6 winners which isn't too bad - but it also turned a tasty profit which is all the better.  Great collects on Richmond, Western Bulldogs and Melbourne (who returned half of the 5.50 odds outlaid) saw a great return for the week.

The round sees the split round, so at the time of writing the predictions below for the first 5 games are based on team form and the players selected, while the last 3 games are based on team form alone and will be updated next week when those squads are announced.

NOTE:  The predictions for the second half of the split round are now updated to include the player's picked.
 

Team 1	Team 2	Venue	Prob 1	Prob 2
Hawthorn	Essendon	MCG	51%	48%
Carlton	Fremantle	Telstra Dome	54%	45%
Brisbane	Richmond	Gabba	73%	25%
North Melbourne	Port Adelaide	Telstra Dome	80%	19%
West Coast	Western Bulldogs	Subiaco	18%	81%
St Kilda	Geelong	MCG	46%	52%
Sydney	Collingwood	ANZ Stadium	51%	45%
Adelaide	Melbourne	AAMI Stadium	41%	58%

AFL 2010 Season - Round 11 Predictions

Last week the model lobbed with 5 winners, as mopst other people would have received as well given it was a week of clear favourites. The good news though is that we managed to get 5 from 5 bets home - including wins on outsiders Essendon @ $3.50 and Brisbane @ $3.73.

This week we've thrown up some interesting picks - including tips on outsiders North Melbourne, Essendon and Western Bulldogs who we rate close to $1.20 chance to beat Collingwood - no doubt we'll see a big bet on them.

Team 1	Team 2	Venue	Prob 1	Prob 2
Richmond	St Kilda	Telstra Dome	20%	79%
Carlton	Melbourne	MCG	54%	44%
Adelaide	Fremantle	AAMI Stadium	28%	70%
North Melbourne	Brisbane	Telstra Dome	52%	47%
West Coast	Geelong	Subiaco	29%	69%
Sydney	Essendon	SCG	46%	52%
Hawthorn	Port Adelaide	MCG	81%	18%
Collingwood	Western Bulldogs	Telstra Dome	18%	81%

AFL 2010 Season - Round 10 Predictions

Back to tipping 6 last week which is a Good sign - the two games we missed out on were Port losing by a point against Melbourne (albeit that were much further down at three-quarter time), and Carlton losing to Hawthorn despite being healthy favourites. Still, we managed a couple of good tips on outsiders in Fremantle away to Sydney, and Adelaide over Brisbane - obviously these were bets as well. Combined with the western Bulldogs who we rated sure things against North Melbourne and it was another profitable week.


This week sees the model agreeing with the bookmakers in just about every game - last night we actually would have had a small bet on Essendon so good signs there. A couple of potential bets include Brisbane at home tonight against Collingwood, and Freo at home tomorrow against North Melbourne.

Team 1	Team 2	Venue	Prob 1	Prob 2
Essendon	Western Bulldogs	Telstra Dome	29%	70%
Geelong	Melbourne	Skilled Stadium	70%	29%
Port Adelaide	Richmond	AAMI Stadium	80%	19%
Brisbane	Collingwood	Gabba	37%	62%
St Kilda	Adelaide	Telstra Dome	83%	16%
Hawthorn	Sydney	MCG	74%	25%
Carlton	West Coast	Telstra Dome	73%	25%
Fremantle	North Melbourne	Subiaco	93%	7%

AFL 2010 Season - Round 9 Predictions

A pretty horrible last week, crashing back to earth with only 3 tips.  And not much better on the betting, though a close call on Richmond certainly would have changed our fortunes.

Not a lot of surprises this wekk - we tip against the favourite in a coupel of close games but otherwise the probs match up with the bookies' odds quite well.

Team 1	Team 2	Venue	Prob 1	Prob 2
Collingwood	Geelong	MCG	49%	50%
North Melbourne	Western Bulldogs	Telstra Dome	10%	90%
Sydney	Fremantle	SCG	47%	52%
Melbourne	Port Adelaide	TIO Stadium	39%	60%
Essendon	Richmond	MCG	76%	23%
Adelaide	Brisbane	AAMI Stadium	53%	45%
Carlton	Hawthorn	Telstra Dome	68%	30%
West Coast	St Kilda	Subiaco	46%	52%

AFL 2010 Season - Round 8 Predictions

It was one of thoseweeks in the AFL last round, we tipped a pretty respectable 6 from 8, including tipping the upset of Fremantle winning away to Brisbane.  As luck would have it though the two bets I had happenes to be on the two tips I missed out on, so no joy on the betting side.

This week sees a few interesting picks - Freo to win quite easily over Collingwood and Brisbane to beat Geelong.  I'll be betting on all home teams, with the exception of the St Kilda vs Essendon game where I'll probably end up with a small wager on the Bombers.

Team 1	Team 2	Venue	Prob 1	Prob 2
Fremantle	Collingwood	Subiaco	69%	29%
Western Bulldogs	Sydney	Manuka Oval	78%	21%
Melbourne	West Coast	MCG	78%	20%
Brisbane	Geelong	Gabba	58%	40%
North Melbourne	Adelaide	Telstra Dome	67%	31%
Richmond	Hawthorn	MCG	32%	66%
Port Adelaide	Carlton	AAMI Stadium	55%	43%
St Kilda	Essendon	Telstra Dome	61%	37%

AFL 2010 Season - Round 6 Predictions

Well we finally sorted the kinks out in the model and tipped favourites in every game in Round 5, only to see half of them lose.  Still, as mentioned in last weeks post we had bets on Melbourne and Carlton who both won at generous odds.

This week sees us tipping all favourites again, except for Essendon over Hawthorn which again I'm comfortable to go with.  So it looks like we'll be betting on Essendon and Richmond, who we rate a lot better than their odds suggest.

Team 1	Team 2	Venue	Prob 1	Prob 2
Western Bulldogs	St Kilda	Telstra Dome	59%	39%
North Melbourne	Melbourne	Telstra Dome	38%	61%
Adelaide	Port Adelaide	AAMI Stadium	42%	56%
Essendon	Hawthorn	MCG	51%	47%
Sydney	Brisbane	SCG	65%	33%
Geelong	Richmond	Skilled Stadium	74%	24%
Carlton	Collingwood	MCG	31%	67%
West Coast	Fremantle	Subiaco	30%	68%

AFL 2010 Season - Round 5 Predictions

Another disastrous week last round, only getting 3 tips right again.  This has seen us revisit the model and tweak it's settings so that it's less dependant on the players selected and moreso on the overall team's form - this would have seen us land 7 tips last week instead of the 3, and 21 overall for the season so far - so I'm happy to go with that.

And thankfully this week we're tipping all the favourites.  We'll probably have a couple of small bets on Melbourne and Carlton which I'm happy enough with - otherwise no surprises at all.

Team 1	Team 2	Venue	Prob 1	Prob 2
Western Bulldogs	Adelaide	Telstra Dome	80%	19%
Sydney	West Coast	SCG	84%	16%
Melbourne	Brisbane	MCG	41%	57%
Port Adelaide	St Kilda	AAMI Stadium	29%	69%
Collingwood	Essendon	MCG	66%	32%
Hawthorn	North Melbourne	Aurora Stadium	69%	29%
Fremantle	Richmond	Subiaco	80%	19%
Carlton	Geelong	MCG	34%	65%

AFL 2010 Season - Round 4 Predictions

Last week was a stinker, only getting 3 games right in the AFL which sees us now languishing in the bottom 25% in most tipping comps. 

This week certainly doesn't do much for my confidence, with the model tipping outsiders in Essendon, North Melbourne and richmond, and also tipping Adelaide by a huge margin in a game the bookies have 50/50.  Still, massive profits were made last year on rounds like this, so let's hope nothing as changed since then!

Team 1	Team 2	Venue	Prob 1	Prob 2
West Coast	Essendon	Subiaco	47%	51%
North Melbourne	Sydney	Telstra Dome	57%	42%
Adelaide	Carlton	AAMI Stadium	86%	14%
Collingwood	Hawthorn	MCG	60%	38%
Brisbane	Western Bulldogs	Gabba	27%	71%
Richmond	Melbourne	MCG	51%	48%
Geelong	Port Adelaide	Skilled Stadium	96%	3%
St Kilda	Fremantle	Telstra Dome	72%	27%

AFL 2010 Season - Round 3 Predictions

Another pretty good week for round 2 in the AFl, with the model getting 6 of the games right and sitting on 12 tips for the season.

This week statrs off inetresting enough, with the model tipping Collingwood in a close game over St Kilda.  Other than that it's pretty much as expected.

Team 1	Team 2	Venue	Prob 1	Prob 2
St Kilda	Collingwood	Telstra Dome	39%	59%
North Melbourne	West Coast	Telstra Dome	65%	33%
Port Adelaide	Brisbane	AAMI Stadium	53%	45%
Sydney	Richmond	SCG	78%	20%
Carlton	Essendon	MCG	73%	26%
Melbourne	Adelaide	MCG	12%	87%
Western Bulldogs	Hawthorn	MCG	65%	33%
Fremantle	Geelong	Subiaco	27%	71%

AFL 2010 Season - Round 2 Predictions

Not a bad week to start off the AFL season, getting 6 of 8 tips correct.

A couple of interesting selections at this stage - Port over West Coast at Subiaco, and Hawthorn in a tight one over Geelong.  Keep in mond the final teams for the last 4 matches haven't been announced yet, so these predictions will change slightly when they are.

Team 1	Team 2	Venue	Prob 1	Prob 2
Brisbane	Carlton	Gabba	63%	35%
Collingwood	Melbourne	MCG	96%	4%
St Kilda	North Melbourne	Telstra Dome	70%	28%
West Coast	Port Adelaide	Subiaco	48%	51%
Adelaide	Sydney	AAMI Stadium	84%	15%
Essendon	Fremantle	Telstra Dome	59%	40%
Richmond	Western Bulldogs	MCG	12%	87%
Hawthorn	Geelong	MCG	49.4%	48.8%

AFL 2010 Season - Round 1 Predictions

The 2010 AFL season proper gets underway today, so too my predictions for each game.  As the model is partly player based, and as only the Richmond vs Carlton teams have been named, I can only publish my predictions for this game at this stage - but as more teams get announced then I'll post the probs up here.

The model has tipped Carlton at a lot shorter price than the bookies have - the reason being that Richmond are playing 3 first-gamers, and the model just doesn't like this.  Even without these players it will still throw the game up as an 80 / 20 chance in Carlton's favour.

Now that teams have been named for Friday and Saturday's game I can release predictions for them - interesting that Geelong are rated shorter and Hawthorn even higher than the bookies odds, all down to the profile of players picked.  The Lions too are rated higher than the bookies so no doubt will be bet on.

A great start last night with Carlton easily account for Richmond, with an easy bet on them collecting.  Also we rated them a 30-point favourite so a bet on them to beat the line of 15.5 points at the start of the game made for another easy collect.

The teams for Sunday's games have now been released and the predictions are now up.  Interesting that it has tipped Collingwood over the Western Bulldogs, based on profile of players selected.  We're currently sitting on 5 from 5 tips so far so let's see if teh form continues.

With some late changes to Port Adelaide's side we now swing our tip over to North Melbourne - it's the cardinal rule of tipping not to change from the original selection and I've done it twice this round already - fingers crossed!!

Team 1	Team 2	Venue	Prob 1	Prob 2
Richmond	Carlton	MCG	13%	86%
Geelong	Essendon	MCG	59%	39%
Melbourne	Hawthorn	MCG	8%	92%
Sydney	St Kilda	ANZ Stadium	36%	63%
Brisbane	West Coast	Gabba	84%	15%
Port Adelaide	North Melbourne	AAMI	42%	46%
Western Bulldogs	Collingwood	Etihad Stadium	45%	53%
Fremantle	Adelaide	Subiaco	35%	64%

AFL 2009 Season - Brownlow Medal Results

An extremely successful year was had this year betting on the Brownlow medal courtesy of the predictions released on this site - thanks of course to Gary Ablett winning at odds of $3 (at least when I placed my bet), massive value when I rated him a 75% chance to win.  The only real downside was Dane Swan not polling as well as I - and most other people expected - this took out a few multiple bets I'd placed, but nonetheless a multitude of other bets out there more than made up for this.

A list of all the bets that returned a profit are listed here:

• Gary Ablett to win - 75% @ $3
• Jimmy Bartel to top group comprised of Cooney, B Harvey, Deledio, Jobe Watson and Scott Thompson - 21% @ $7.5
• Bernie Vince to top group comprised of Adam Selwood, Rich, Akermanis, Hodge and Hasleby - 66% @ $2
• Fevola to top group comprised of Koschitzke, Franklin, LeCras, Pavlich and Tredrea - 75% @ $3
• Vince to top Adelaide - 53% @ $2.40
• Brown to top Brisbane 82% @ $2.2
• Winderlich to top Essendon - 17% @ $34
• Pavlich to top Fremantle - 57% @ $4
• Ablett to top Geelong - 99% at 1.12
• Richardson to top Richmond - under Any Other Player (not Deledio, Foley, Tambling, Jackson or Tuck) 49% @ $4
• LeCras to top West Coast - 34% @ 3.25

A couple of these options ended up tying with another player, effectively  halving the odds - having said that we managed to return over 50% profit on the night mainly thanks to Ablett winning and Winderlich topping Essendon at long odds.

A couple of missed opportunities were there as well - the first was Jonathan Brown to finish in the top 5 overall.  I rated him 5th in my predictions, which is exactly where he placed at massive odds of $18.  

Another was Dane Swan - he was paying $1.03 to top Collingwood's votes, whereas I rated him 91% chance, so under the odds - this represented a very nice laying opportunity where you could have taken on someone else's massive bet for a very minor risk - ie if you took $1,000 of someone's money, you would have only risked paying out $30 had Dane Swan topped Collingwood.  Turns out he didn't, so you would have pocketed the $1,000.

Still it was a very successful event overall, and in the days since I've come across a plethora of detailed player statistics that should only improve the model and it's results.  And as always, if not well we still have a model that is extremely capable.

So the only piece of AFL prediction left for the year will be the grand final, which I'll upload after the final team selections are announced.  At this stage on team form alone I'd rate Geelong slightly on top, however once we factor in the player's named this may change.  Til then.

AFL 2009 Season - Brownlow Medal Predictions

Every year for the past few years I've compiled Brownlow Medal predictions with reasonable success.  Every game I rate each player on their chance to get 3, 2, or 1 votes - then based on these probabilities I simulate the season 30,000 times in order to find the average number of votes for each player over the whole season, and their respective chance to win the medal. 

A variety of statistics are fed into the Brownlow Model - for example, each player's kicks, marks, handballs, goals, inside 50's and even the number of free kicks they give away are found to be significant predictors of how many votes they will receive.  However the two most significant predictors are the margin your team wins by, and how many Supercoach points you receive in a game.  The latter is by far the major pointer to how well a player is likely to poll, and explains why I rate Gary Ablett so highly this year, as he has by far amassed the most Supercoach points per game played.

Given that the entire season is simulated, it enables me to produce a variety predictions, not just who will win the league.  Other stats I've produced include the likelihood of there being multiple winners, how many votes the winner is likely to poll, various head to head and group propositions, and who is likely to poll the most votes for each of the teams.

One point worth noting - these predictions are all based on cold, hard, observable statistics.  There's no accounting for other reasons a player might poll well - such as being based in the midfield, having blonde hair, or being named Adam Goodes for example.  All other things being equal, a midfielder kicking for goals will likely poll higher than a forward kicking 4 goals.

So here are the predictions for this year - if there is anything else you'd like to see let me know and I should be able to work it out.  For what it's worth I rated last year's winner Adam Cooney a 15% chance to win - he paid $14 so this represents close to double what I rated him, good value indeed

Click on the link to download a spreadsheet of advanced predictions

2009 Brownlow Medal Predictions

Player	Avg Votes	% Chance
Gary Ablett	26.8	75.08%
Dane Swan	22.4	21.93%
Nick Dal Santo	18.2	4.84%
Leigh Montagna	17.8	3.56%
Jonathan Brown	15.9	0.85%
Brendon Goddard	14.1	0.64%
Paul Chapman	15.1	0.38%
Sam Mitchell	13.3	0.00%
Alan Didak	14.4	0.27%
Lenny Hayes	12.3	0.19%
Chris Judd	12.4	0.18%
Nick Riewoldt	12.1	0.17%
Marc Murphy	11.5	0.13%
Adam Goodes	12.5	0.06%
Bernie Vince	10.6	0.05%
Brendan Fevola	12.4	0.03%
Luke Power	10.2	0.03%
Matthew Boyd	9.5	0.03%
Joel Selwood	10.0	0.02%

AFL 2009 Season - Finals Week 3 Predictions

Apologies for the late posting of these, I've spent most of my free time this week getting my Brownlow Medal predictions handy, which I'll post on here shortly.

Well first off last week went according to script, firstly with the Western Bulldogs winning easily and Collingwood winning in a very close match.  Both these teams provided decent overlays as well, so profits were to be made for those betting.

I'm pretty hesitant to put the predictions up for this week, and look at the probabilities you'll soon see why, tipping both underdogs by a fairly big gap.  As has been the case with the Western Bulldogs in recent weeks, on form alone we'd rate St Kilda and Geelong on top but when the player-based aspect is added things swing in favour of the underdogs.  For what it's worth I'll be extremely happy if both of my tips are wrong, as this means St Kilda will be in the Grand Final, and more importantly Collingwood won't be.

Team 1	Team 2	Venue	Pr 1	Pr 2
St Kilda	Western Bulldogs	MCG	26.6%	71.7%
Geelong	Collingwood	MCG	20.2%	78.4%

Total tips for the season: 128 from 182 (70%)
Total tips for finals: 5 from 6 (83%)

AFL 2009 Season - Finals Week 2 Predictions

We tipped 3 out of the 4 winners in the 1st week of the final - missing out the Western Bulldogs who I rated highly - which is the case again this week, in giving them a 92% chance against Brisbane, and again because of the profile of players selected (we'd still tip them just based on both teams' form alone for what it's worth).  The other game I rated almost a toss of the coin, tipping just in favour of Collingwood.  Any further teams changes would be interesting as this could sway the predicition in Adelaide's favour.

As a said last week I'm not using my predicstions to bet on the finals, primarily because the data I used to build the model is only using games from the home-and-away rounds.

Team 1	Team 2	Venue	Pr 1	Pr 2
Western Bulldogs	Brisbane	MCG	91.7%	7.7%
Collingwood	Adelaide	MCG	49.5%	48.4%

Total tips for the season: 126 from 180 (70%)
Total tips for finals: 3 from 4 (75%)

AFL 2009 Season - Finals Week 1 Predictions

5 tips in Round 22 to round off the home-and-away season, for a final figure of 123 tips (69.9%) overall which is quite respectable for a start-up model.  As I mentioned in the season review we should only see this figure to increase when I improve the model in the off-season.

In the meantime here are my predictions for week 1 of the finals series.  No surprises for 3 of games, obviously the Cats vs Dogs game is an eye-opener with the Dogs rated so high.  Again this is down to the profile of players selected, a couple of nuances in certain players profiles sees the massive probability here.  We rated the Dogs highly in their last two game against more fancied opposition and they managed to win these, so again will be interesting to see how this game pans out.  For the record I don't think I'll be betting heavily on this game, or indeed through the finals.

Team 1	Team 2	Venue	Pr 1	Pr 2
Adelaide	Essendon	AAMI Stadium	79.3%	19.4%
Geelong	Western Bulldogs	MCG	7.9%	91.4%
Brisbane	Carlton	Gabba	78.2%	20.4%
St Kilda	Collingwood	MCG	68.7%	29.6%

Total tips for the season: 123 from 176 (69.9%)
Total tips for finals: N/A

AFL 2009 Season - Regular Season Review

Now that the home-and-away matches have been completed I thought I'd share the overall results of the model for the year, both in terms of tipping and betting.

Tipping

The tipping did quite well for a first attempt at an AFL model - finishing on 123 tips at 69.9% success (counting both draws as 0.5 tips). The leading Herald Sun football expert finished on 122 tips, so we topped them, while Shane Warne finished on an impressive 128, no doubt buoyed by tipping St Kilda every week.

Comparing against the results from footytips.com.au (who count a draw as a whole tip, not just 0.5) - out of the 317,000 tippers we would have finished in the top 5% - 7% which again is a pretty good result for a start-up model. Hopefully some improvements to the model will only improve on this figure - more on this later.

Betting

Tipping is nice and fun and goes a long way towards bragging rights, but if a model can be used to make consistent profits then all the better. As mentioned in previous posts I follow the Kelly Criterion method of betting - which has the useful property of maximising one's exponential profit. The criteria works by finding out if the model throws up an advantage over the bookmaker (for example, if a team is rated 50% and a bookie is offering $2.20 then we have an advantage, if they are only offering $1.80 then we don't). The higher the advantage over the bookie, the more that is bet.

Two options are presented in the results below. The first shows the results had we potentially bet on every game, provided that there was an advantage over the bookie. The second option will only bet on a team if they are playing in their home state, or at a neutral venue. So in this scenario if Hawthorn played West Cost in Perth and provided a betting opportunity, we wouldn't bet on them. However if these teams played in Tasmania, then we would bet on Hawthorn as it is a neutral venue. Respected sports predictions website Sportpunter has shown in the past that betting on teams travelling interstate isn't as profitable as those playing at home or neutrally, and in fact may end up losing money. So we will present both scenarios and see if there is a significant difference.

The analysis that follows will compare various betting strategies dependent on how much we rate a team to win, and how much of an bookie's advantage (overlay) we have.

The first piece of analysis we'll look at here is betting on every game, provided there is a an advantage to be found. So regardless of how highly we rated a team, or the size of the overlay we'll bet on them - so long as there is an overlay. As mentioned, we'll compare betting on all games to just betting on home/neutral teams.

Return on Investment

• All games: 17%
• Home/Neutral Only: 25%

The return on investment is simply the amount profited (or lost) as a percentage of the total amount bet. So looking at the above graph we can see that only betting on home/neutral teams returned around $1050 profit while betting on all games returned around $900 - instantly we can see in this scenario that the bets on the teams playing interstate actually lost $150 - so the second option is better here as less money is bet (as we are not betting on interstate teams), yet profited more.

The next scenario we'll look at is to only bet on teams provided there is an overlay, and we rate them a least a certain percentage to win - in this case we'll look at 60%


Return on Investment

• All games: 25%
• Home/Neutral Only: 29%

So in this case both scenarios return around $1,100 profit, but remember that by excluding those team playing interstate we are in effect risking less money to win the same amount (hence the higher ROI).

Finally we'll look at a scenario where we only bet on a team if they have at least a certain overlay over the bookmaker. Again, any figure can be chosen - for this example we'll look at a minimum 34% overlay before we bet on a team


Return on Investment

• All games: 44%
• Home/Neutral Only: 60%

These scenarios return a little less than the previous ones, showing about $850 profit. However with this filter we have risked only half as much money as before - resulting in the massive ROI figures shown above. Also when looking at the home/neutral scenario this appears to have a nice straight-line trend throughout the year, whereas the other scenarios - while constantly growing throughout the year - appear to tail off towards the end.

Regardless of the combination of scenarios chosen, it is clear to see this model has performed exceptionally well in betting over the 2009 season.

Improvements

Despite the success of this model, there is a lot that can be improved on, which will hopefully see an uplift in both tips and profitability.

1. A portion of the model is player-based, hence I need to obtain the players that are selected on a Thursday night for that weekend's games. While I try and update the teams if any late changes are announced, for any number of reason the list of players I obtain may not be the final 22 who run out on game day - particularly for Sunday games. By going back and entering the actual players who played each game we should expect to see a slight improvement in the model's performance.
2. The player-based aspect of the model is fairly simple - I only look at the number of games played and goals kicked for the player's selected, as this info was easy to find and incorporate into the model at the time. However there is a wealth of player-based information available - for example disposals, effectiveness, time on ground, even player ratings such as SuperCoach points - so I will see if adding this information into the model will improve it significantly.
3. This model is built only using data from the 2007 and 2008 seasons, and applied to the 2009 season to arrive at the results shown on this page. By going back a few more years we should be able to bring more information into the model and hopefully improve upon the results.
4. At the moment the home ground advantage in the model is a generic one applied to the whole league. For example, all other things being even, a Melbourne team would have just as much advantage over a South Australian team as Brisbane would over a team coming from Perth. It may be beneficial to look at home ground advantage based on distance travelled, or even give each team their own home ground advantage.
5. Given the scenario above where betting on a team playing interstate isn't profitable, I may look at building two separate models - one model based on both teams being in their home state or at a neutral venue, and one where there is a team in their home state playing a team from interstate. Whether this makes a difference or not I have no idea, but the results will be interesting to see.

So there we have it. I'll still aim to publish predictions for the finals but it is worth noting that these models are only built using data from the home-and-away season so I give no guarantees for the finals. Also in the next couple of weeks I hope to publish my predictions for the Brownlow Medal.

There's probably a lot of foreign concepts introduced here so if anyone has any queries at all please don't hesitate to leave a comment and I'll get back to you.

AFL 2009 Season - Round 22 Predictions

Finally a good week for the AFL tips since I started publishing them - tipping 7 out of 8 correct games, the game I missed being 1st place St Kilda losing to 13th placed North Melbourne. Still, some of the unusual predictions the model threw up were justified, the Western Bulldogs over Geelong in particular.

So here are the final tips for the home-and-away matches for the year, quite a few 50/50 propositions, and again it has tipped the Western Bulldogs as huge favourites in a game the bookies have Collingwood as slight favourites.

Team 1	Team 2	Venue	Pr 1	Pr 2
West Coast	Richmond	Subiaco	50.2%	47.7%
Geelong	Fremantle	Skilled Stadium	64.3%	33.8%
Hawthorn	Essendon	MCG	52.4%	45.5%
Carlton	Adelaide	Telstra Dome	45.8%	52.1%
Sydney	Brisbane	SCG	70.1%	28.2%
Port Adelaide	North Melbourne	AAMI Stadium	71.9%	26.5%
Melbourne	St Kilda	MCG	2.3%	97.5%
Collingwood	Western Bulldogs	Telstra Dome	14.4%	84.5%

Total tips for the season: 118 from 168 (70.2%)

AFL 2009 Season - Round 21 Predictions

Murphy's Law still kicking around - only managing to tip 3 winners last week, combined with the 2.5 winners from round 19, when the lowest round I had prior to making these tips public was 4.

Anyway we soldier on, though looking at the first game has cause to make me nervous, having the Western Bulldogs as an 85% chance against Geelong - mainly due to the profile of players selected. The player-based aspect of the model will be getting a serious overall in the off-season, but as it stands it is sitting on 111 tips so far, so hopefully this game is an exception rather than the rule. This game aside, the other predictions look quite reasonable.

Team 1	Team 2	Venue	Pr 1	Pr 2
Western Bulldogs	Geelong	Telstra Dome	84.9%	14.0%
Carlton	Melbourne	Telstra Dome	87.2%	11.9%
Adelaide	West Coast	AAMI Stadium	89.1%	10.2%
Brisbane	Port Adelaide	Gabba	65.8%	32.4%
Richmond	Hawthorn	MCG	20.3%	78.3%
St Kilda	North Melbourne	Telstra Dome	92.3%	7.1%
Collingwood	Sydney	MCG	72.2%	26.2%
Fremantle	Essendon	Subiaco	53.5%	44.5%

Total tips for the season: 111 from 160 (69.4%)