Parker's Pythagorean Playbook

Round 8 Tips Carlton v Brisbane (43%, 13pts) Richmond (78%, 24pts) v West Coast Geelong (75%, 16pts) v Adelaide Gold Coast v Melbourne (66%, 16pts) GWS v Western Bulldogs (68%, 20pts) Fremantle (69%, 13pts) v Hawthorn Port Adelaide (58%, 7pts) v Essendon Collingwood (68%, 10pts) v Sydney North Melbourne v St Kilda (84%, 40pts) Not really many surprises with the favourites but I have tweaked the margin predicting on the model to provide more realistic results based on how the ground impacts scoring and the rest teams have had coming into the game. Head to Head Bets Season performance so far: 4.35 bits, 27.69 MAE, 39 tips (62%), -2.02 units, -10.4% ROI Three bets this week, one favourite, one underdog and a coin flip. Brisbane at $1.95 against Carlton - 1 unit. This is a really surprising market considering Carlton have been ordinary so far and Brisbane have showed they can turn it on when they want to. Fair odds for Brisbane is $1.68 so interested to see how this plays out. Western Bull...

How Did the Model Perform? Nearly every model was in consensus on every game but the footy gods spoke and we got a few upsets in round 7. The PPP model went with the crowd on all the matches and tipped every favourite, resulting in 6/9 tips thanks to Port Adelaide, GWS and Gold Coast all upsetting their more fancied rivals. Bits We gained some more bits with some confident projections coming true, up to a new season high of 4.35, 0.85 gained this round alone.  We have moved up a spot on the Squiggle standings to 22nd place, looking to continue the upward trajectory moving forward. MAE We lost ground on MAE with some big thrashings of teams being a little outside the predicted margins. We move to 27.69 on the year with the round average of 30.91.  We move to 23rd on Squiggle for MAE with most of the models now becoming more correlated with MAE to bits as we move into the season. I have made some edits to the margin prediciting of my model which has been backtested to be about 0...

Round 7 Tips St Kilda (64%, 14pts) v Port Adelaide Brisbane (69%, 20pts) v Fremantle Sydney (78%, 31pts) v GWS Western Bulldogs (72%, 23pts) v Hawthorn Melbourne (85%, 43pts) v North Melbourne West Coast v Carlton (70%, 21pts) Essendon v Geelong (62%, 12pts) Richmond (59%, 9pts) v Gold Coast Collingwood (60%, 10pts) v Adelaide Suggested Bets for Round 7 Season performance so far: 3.50 bits, 27.15 MAE, 33/54 tips (61%), -4.02 units, -21.7% ROI Gold Coast at $2.90 to beat Richmond - 1 unit This game for Richmond is massive alarm bells. A loss here and they drop to 1 win of 7 matches and a skid that has no end in sight. They got up for the match against Melbourne on "ANZAC Day Eve" but with this match being a Sunday afternoon meaningless time slot, there are worrying signs at Tigerland.

Where is the Model Looking? Similar old story for the model after week 6 with a lot of expected teams winning and no one (apart from Sydney) getting unexpectedly blown out. We still like the looks of St Kilda and Essendon more than the market while laying Richmond and Carlton, with the Western Bulldogs coming back into normalcy after 6 weeks of action. Futures Look Ahead We have chosen to go a bit deeper into some action for this week's picks with all bets already having been bet on a number of times in the first five weeks. There are four bets to have a look at this week, all revolving around making and missing finals. Essendon finals - $2.10 ($1.63) 0.5 units Western Bulldogs finals - $2.10 ($1.98) 0.5 units Adelaide finals - $1.95 ($1.78) 0.5 units Carlton miss finals - $2.15 ($1.75) 0.5 units This moves our season total up to 31.25 units.

How Did the Model Perform? The easiest round to tip on the year to date, round 6 saw us grab 8 of the 9 available matches correctly, the only blemish coming on Friday night's game where the Doggies got the better of the favoured Dockers. Bits Coming into this round we had a total of 1.01 bits from the first 5 rounds combined. After the latest slate of matches we are now up to 3.50 bits, up 2.49 in round 6 alone! Our tips however were a bit less confident than some other models and we now trail first place AFLalytics by nearly 5 full bits. We sit 23rd on the Squiggle leaderboard, down one place from last week. MAE Another fantastic result for MAE this round with 23.65 average, just slightly worse than the 23.60 from round 5. On the season we improve to 27.15, thanks to an exact 13 point margin on the ANZAC Day match, saluting after the siren. On Squiggle we sit in 25th for MAE, down three spots from last week. However we are only 1 MAE off a top 10 spot, plenty of work still to be d...

Round 6 Tips Fremantle (59%, 9pts) v Western Bulldogs Port Adelaide (79%, 34pts) v West Coast GWS v Brisbane (69%, 20pts) Geelong (50.3%,1pt) v Sydney Hawthorn v Adelaide (64%, 14pts) Carlton v St Kilda (55%, 5pts) Gold Coast (71%, 22pts) v North Melbourne Melbourne (62%, 12pts) v Richmond Collingwood (63%, 13pts) v Essendon Suggested Bets for Round 6 Season performance so far: 1.01 bits, 27.85 MAE, 25/45 tips (56%), -2.3 units, -18.4% ROI Sydney at $2.63 to beat Geelong, 2 units Hawthorn at $3.80 to beat Adelaide, 1 unit St Kilda at $2.14 to beat Cartlon, 2 units Richmond at $3.55 to beat Melbourne, 1 unit A massive 6 units of wagers placed this week on the four selections. We think the HFA at GMHBA will be big but not enough to suggest a $2.60 or above price on a good Sydney team. St Kilda probably should be favourites against an out of form Carlton and Richmond are more than capable on their day to take down the now shaky Melbourne if they hit another off day.

Where is the Model Looking? Some good results for our model in round 5 meant we are in a good spot for our futures bets up to this point of the season. If we assume any current odds of <2 as a "win" and any >=2 as a "loss", we can say we are up on the year 1 unit on our bets. Of course there are still 19 rounds of football to go and any team can finish in any position, but it is a good base to build from. The model likes the usual suspects this round again, St Kilda, Essendon and Adelaide, but the suggested bets aren't that much different to previous weeks, resulting in a slower week for punting. Futures Look Ahead There are many double and triple up bets we could take going into round 6 but this feels like an overkill with so many wagers concentrated around the same markets as it stands. We could go again on Richmond missing finals but we have 4.25 units there already. We could have a go at Carlton or Port Adelaide missing the 8 but again we have 1.5 unit...

How Did the Model Perform? A relatively straight forward round 5 allowed the model to make up some much needed bits and saw the MAE reduce down to a season low. The MAE for this round was a lovely 23.6 and we gained 1.01 bits over the 9 games. However as this was the case for most models, it meant we did in fact lose ground compared to some of the Squiggle models who had more confident picks for the winning games. A good result for the model on futures betting markets but a bad result for season performance ratings, a very mixed bag. Bits As mentioned, we improved to a season high 1.01 bits, thanks to a 1.01 bit weekend. Geelong gave us the biggest improvement of 0.67 bits and Brisbane winning was earned us a 0.66 return but we instantly gave back 0.54 bits when Melbourne went down to Essendon. Interestingly we would have profited or lost 0 bits no matter who won the GWS v Hawthorn game, our MAE of 2.01 was extremely good but it ended up being a net negative for the model compared to o...

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 ...

Gather Round (Round 5) Tips Adelaide v Carlton (57%, 7pts) Fremantle (56%, 6pts) v Gold Coast Richmond v Sydney (57%, 7pts) Brisbane (79%, 33pts) v North Melbourne Essendon v Melbourne (66%, 16 pts) Port Adelaide (53%, 3pts) v Western Bulldogs  Geelong (79%, 33pts) v West Coast GWS v Hawthorn (50.0001%, 0.0006pts) Collingwood (51%, 1pts) v St Kilda Wow. A lot of these games are a lot closer than I would have initially predicted. Carlton and Fremantle are both worthy favourites although I would have thought both would be 5+% more confident. Sydney being a 57% favourite over a depleted Richmond is massively under the odds for me. At first glance I had this in 60-65% Sydney range, I guess the loss to Port Adelaide did a lot of harm to their rating. Collingwood also probably should be a little more favourites based on what we have seen this year but the model only knows a win or loss and by how much and St Kilda are 4-0 so there is evidence there to support them. You can only beat who ...

Where is the Model Looking? Round four has thrown up some results that have meant our ratings on certain teams and the bookmakers' are once again quite different. The teams that are loved by the model seem to be St Kilda, the model has them favourites for the minor premiership at 22% and it also likes Essendon, PPP has them with a 56% finals chance compared to the bookies' 28%. Compare this to the other end and the market as a whole isn't quite sure on Adelaide, some models have them 10% for finals, some as high as 45%, ours is slightly bullish on 35%. Geelong also are rather confusing with some models very confident and some thinking 6th or 7th is their position. Futures Look Ahead So where should we be betting for the final ladder positions? Six bets in total however one of them is a whopping 3 unit bet with the bookmakers throwing up a wrong price.  A few of these bets counteract some earlier season bets, however as our ratings fluctuate a little more than the bookies...

How Did the Model Perform? It was a mixed bag, round four, and it did not treat the PPP model kindly. A mediocre 5 tips correct with some pretty big losses thrown in there means it is likely to be our worst round of the young season.  As I wrote last week , this round meant a lot to the model thanks to some polarising predictions and we did not get the chocolates for the Easter weekend, losing all four of the "50/50" games. Adelaide running over Fremantle meant we forewent 0.24 bits, Brisbane winning cost us 0.38 bits, Bulldogs another 0.33 and Port Adelaide a whopping 0.76. Those go the other way and we are near the top of the ratings but as luck would have it we missed the mark this week and are now chasing our tail. For the entire round we did profit bits, up to a dead even 0.00 now after winning 0.75 bits cumulatively. Our MAE declined ever so slightly by 0.06 to 28.94 and we look to improve going into round five. Bits We dropped a few places on the  Squiggle  leaderb...

Round 4 Tips Brisbane v Collingwood (62%, 12pts) North Melbourne v Carlton (74%, 26pts) Adelaide v Fremantle (58%, 8pts) Richmond (60%, 10pts) v Western Bulldogs St Kilda (66%, 17ts) v Gold Coast Sydney (70%, 22pts) v Port Adelaide Essendon (65%, 15pts) v GWS West Coast v Melbourne (78%, 31pts) Geelong (70%, 21pts) v Hawthorn The only game here I was surprised to see is Richmond favourites over Western Bulldogs, I would have thought the Doggies' win on Thursday night would have been enough to give them the edge here but Luke Beveridge's men start with a 58.9 elo point deficit and then after HFA is added the Tigers come out ahead. This round is the largest average of confidence with only one game in the 50-60% range and four above 70%. This means it is a massive round for bits to be won or lost so expect lots of movement on the bits leaderboard. Suggested Bets for Round 4 Season performance so far: -0.75 bits, 28.78 MAE, 14/27 tips (52%), +0.2 units, 3.1% ROI Only the two bets f...

Where is the Model Looking? For most teams, the PPP model and the bookmaker's markets are converging to one standard opinion. This bodes well for us as we know sometimes models can be far outside the range of "reality". No one really knows what the future holds but the bookies are historically somewhat accurate so we trust them as a reliable source for future telling. Our model could be spot on and the bookies way off, although this is rather unlikely. There are however a few teams defying the group think and forging their own path in the PPP futures markets.  Coming off three impressive wins, St Kilda are the model darling, some 25 or more percent chance better to make the finals than the markets suggest. This is helped out by beating some what strong teams and doing so while being underdogs every time. Essendon for similar reasons are also loved by the model. Note their loss to St Kilda does not impact them as much as it probably should because we think St Kilda is also...

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...