2013 : Round 6 - Results

As Investors will know, the Line Fund carries a 60% weighting in the Recommended Portfolio, so when it ended Saturday with an 0 and 4 record, knocking 6c off the Portfolio's price in so doing, Sunday was always going to be a day about, at best, limiting the damage. 

And limit it we did, eventually shedding just 1.5c from the Portfolio across the entire weekend thanks to three successful Sunday line bets, diluted just a little by 6 unsuccessful SuperMargin wagers on the same day. Three of Sunday's SuperMargin wagers were genuine chances until late into their respective games, but it's hard to feel too aggrieved by their failure when you balance them against the two successful SuperMargin wagers from Saturday, each secured by a single point. (Still, Sydney, did you really have to score that last behind?)

To summarise, the Head-to-Head Fund landed 0 from 3, dropping 1.6c, the Line Fund landed 3 from 7, dropping 3.3c, and the Margin Fund landed 3 from 12, gaining 9.4c. That dropped the Recommended Portfolio by 1.5c leaving it down 3.2c on the season.

Two of those successful SuperMargin wagers came, once again, from Combo_NN2 predictions, which means that this Predictor has selected the correct bucket in 12 of the season's 54 games, and been in error by a single bucket in 7 more games.

That 70.4% figure recorded in the CN2 row means that, had someone level-stake wagered on all of Combo_NN2's SuperMargin predictions, they'd be showing a 70.4% return on their wagers so far this season.


The incredibly predictable nature of the season continued this week, with all nine favourites collecting the four competition points and many of MAFL's Head-to-Head Tipsters, Combo_NN2 included, tipping the card. That leaves Combo_NN2 on top of the Tipping Leaderboard, still one tip clear of the field.

Favouritism has dominated Home team status this year like no other in recent memory: favourites have won over 74% of contests while Home teams have won only 50%.

As I think I've noted before, it's not only the names of the winners that have been relatively easy to predict this year, but also the margins of their victories. Even the worst-performed Margin Predictor this week turned in a mean absolute prediction error (MAPE) of just over 20 points per game, while the best MAPE belonged to Combo_NN2 and was a tiny 15.8 points per game. I can't help but feel that at some point, the season will receive a large injection of entropy. Here's hoping it's via attractively-priced and astutely-detected Home team underdogs.

The two RSMP-based predictors also made extremely accurate margin predictions this week and, in so doing, retained the top two places on the Leaderboard. They also have the highest percentage of correct line market predictions so far this season (63%).

Amongst the Head-to-Head Probability Predictors, the three TAB-based Predictors continue to hold down the top 3 spots and retain the same order as they held at the end of the previous round. The Line Fund algorithm recorded its worst single-round performance this year and so still has a slightly negative average probability score for the season to date.

Next week the Head-to-Head Fund's wagers are doubled, so Investors will need to hope for a marked improvement in this Fund's ability to identify positive-expectation wagers.