2015 - Round 11 Results: Suns Almost Shine Over Fre

Again this week it fell to a single team to provide the only bright spot in an otherwise dark and depressing round of wagering. It was the Pies who eventually gave that relief when they comfortably accounted for the Giants on Sunday, though the Suns, which we'd bought at $10, had gone very close to lighting up the entire weekend only 24 hours earlier.

The Suns' narrow loss was added to later on Saturday by another, almost as narrow loss by the Roos, the net result of those two losses, even offset as they were a little by the Pies' Sunday win, seeing the Head-to-Head Fund slip another 1.7c to be down by 10.8c. 

The Line Fund fared no better, registering its fourth successive loss and its ninth in 11 rounds after Port Adelaide comprehensively failed to cover its 13.5 point spread on Friday night. The Fund shed another 2.5c as a consequence and is now down by 15.4c on the season. It and the Head-to-Head Fund are therefore now down, in Portfolio-weighted terms, by 13.6c.

(By the way, if you're wondering about the reference in this week's blog title, see this.) 

TIPS AND PREDICTIONS

Only three home teams but five favourites won in Round 11, which was bad news for Home Sweet Home and led to its equal-worst three from six performance, but generally good news for the Head-to-Head Tipsters as a group, since they collectively averaged 4.4 correct tips from six.

BKB, Bookie_3 and Bookie_9 all registered five from six performances to take their season records to 67 and 29, which is an accuracy rate of just under 70%.

View fullsize

MoS' Margin Predictors registered an all-Predictor Mean Absolute Error (MAE) of 23.6 points per game this week, their second-best result for the season, behind only the 23.3 points per game result they produced in Round 9. Best was RSMP_Simple, whose 18.7 points per game MAE lifted it from 5th to 3rd on the MoS Leaderboard where it trails RSMP_Weighted, whose MAE of 19.2 lifted it from 3rd into 1st, and ENS_Greedy, whose MAE of 22.8 was enough to preserve its 2nd-place ranking.

ENS_Linear was the big faller this week among the Margin Predictor elite. It slipped from 1st to 5th after an MAE of 24.6 points per game. It's still though less than 21 points behind the leader.

View fullsize

It's interesting to note that, currently, five of the six best Margin Predictors are ensemble-based predictors, and the sixth is an ELO-based predictor that completely ignores bookmaker opinion. If you'd offered me that situation at the start of the season, though I might have asked for C_Marg to be a little higher on the ladder, I'm fairly certain I'd have taken it.

In the world of Head-to-Head Probability Predictors the news is also good, the all-Predictor average finishing positive for the fifth successive round. That's to be expected, of course, when favourites have been winning as frequently as they have. This week's best probability score was registered by Bookie-RE, though its score was effectively equal to Bookie-LPSO, which still occupies first place on the Leaderboard. C_Prob continues to register positive but unexceptional probability scores and sits snugly in fourth position, a safe distance above the pursuing, older Predictors.

Finally, the Line Fund algorithm, in what Investors might hope will be a positive sign for the prospects of the Line Fund over coming weeks, registered its second positive probability score in three weeks, and only its third for the season.