For a fleeting few minutes on Saturday afternoon as the Dogs seemed destined to secure the most remarkable of comebacks, MAFL was notionally in profit. But the Eagles rebounded, kicking the last 14 points, compounding Investor disappointment by not just winning but by covering the spread as well. Then, later that night, Freo capitulated to the Hawks, leaving it all up to the Pies and the Crows to salvage the round.
Fortunately that's exactly what they did, though not before the Pies had effectively spotted the Dons 77 points start by adding a 30-point scoreboard deficit to the 46.5 points spread they'd offered - and we'd granted - pre-game. A 16 goal to 3 second-half from the Pies proved sufficient to grab the competition points, to secure both our head-to-head and line bets, and to shore up the Pies' credentials as competition favourites.
At the conclusion of festivities/hostilities (you choose which), the Head-to-Head Fund was down another 9% and the Line Fund was up almost 3%, in total knocking just over 3c off the price of Investor Portfolios.
Across the round's seven contests the Eagles were the only favourites seriously tested, and their eventual victory made for another clean-sweep by the bookmaker's preferred teams this week. As a result the average Head-to-Head Tipster performance for the round was a high 5.9 from 7, which is the second-highest accuracy rate for the season so far.
Tipsters at the top of the leaderboard all tipped the card so there's been no change in their rankings and just a small increase in their already-impressive season-long tipping percentages. Bookie_9 and Combo_NN_1 now lead out on 111.5 from 147 (76%).
When a team wins by 233 points as the Cats did on Saturday - incidentally securing a 0.04c increase in Investor Portfolios - you just know it's not going to be a good week for margin prediction, and this week's average of 39.25 points per game, the second-worst all-Predictor average this season, bears that out.
Combo_NN_1 tallied the round's best MAPE at 33.66 points per game, ahead of Combo_7 on 35.67 points per game. Amongst the other strong Margin Predictors, Bookie_3 was next best at 36.35 points per game, then Bookie_9 at 38.27 points per game, and Combo_NN_2 at 39.04 points per game.
That leaves Bookie_3 still atop the MAPE leaderboard with a season-long MAPE of 28.33, but has allowed Combo_7 to narrow the gap between it and Bookie_3 to just 0.54 points per game. Bookie_9, in third and now on 29.52 points per game, is the only other Margin Predictor currently with a sub-30 MAPE.
Predicting line winners was also difficult this week. The all-Predictor average of just 2.4 from 7 is the third-lowest average for the season (adjusting for the fact that this was a short round), reversing the trend of recent weeks of above-average line betting performances.
Head-to-head probability scores were very high, however, highest of all for the TAB Bookmaker who extended his lead over H2H Adjusted and Unadjusted on the Head-to-Head Probability Scoring table. The Line Fund algorithm recorded another slightly negative probability score this week.