2011 Round 13: A Little Respite

Investors can look forward to a much quieter round this week - we've got a whole game to watch bet-free and three, arguably four more games that don't matter much whichever way they finish.

The diminished activity has two sources. Firstly, the Head-to-Head Fund has found only two Home teams that it fancies: St Kilda, facing Geelong, which it likes just a little, attracted presumably by the $3.75 on offer; and Carlton, facing the Swans, which it's downright infatuated with at $1.25.

Secondly, the Line Fund has only five wagers this week. It wanted three more but, while performing an analysis of the Line Fund's performance for an upcoming blog on the Statistical Analyses journal, I was struck by the historical and current-season asymmetry in the Fund's profitability when wagering on Home teams, where it's been profitable, and Away teams, where it hasn't. That led me to review this blog from earlier in the season where I'd noted that the Line Fund would wager only on Home teams this year, a conclusion I'd reached after a thorough and careful analysis of the Line Fund's performance during which I'd noted the asymmetry in ... well, you get the picture. Anyway, the Line Fund's been wagering on Away teams for 12 rounds in complete breach of its Terms & Conditions, which stops as of now. I'll be enforcing the no-Away team wager rule for the Fund from this round onwards. 

(For the record, the Line Fund would have had 0.5% on Richmond, 2% on Geelong, and 6.4% on Sydney this week had I not prevented it. We can now all share in the joy of a rule well-enforced or the regret of a rule enforced-in-error as the case may be.)

Wagering on Away teams has cost the Line Fund just under 6% this season, an amount which as of now I'm reinstating, consistent with the MAFL Tradition for handling screwups (try a site search on Special Stupidity Dividend and find the relevant Newsletters for more details). As a result Investor Portfolios rise by about 3% to just under 87c.

Here's how it now looks: 

With the newly-enforced Line Fund prohibitions now in force, the week's wagering activity looks like this:

As indicated, only three games are of much import then. If the Saints win, Investor Portfolios rise by 3.7%, which is the largest gain that a single game can deliver this week. A Carlton win is worth another 1.9%, but a Swans victory instead would cost almost 8%. Lastly, a Dons win by 9 points or more would add 2% to Investor Portolios, while a Roos loss would subtract roughly the same amount.


To head-to-head tipping then where we find four games with unanimous support for the favourites and three more with majority support by at least 8-5 for the favourites. Only in the Melbourne v Fremantle game is there majority support for the underdogs.

Next, margin prediction, where the majority support of the Margin Predictors is the same as that for the Head-to-Head Tipsters in all eight games. The only game for which there are fewer than 11 tipsters selecting the same team is in the Melbourne v Fremantle clash where there's 9-4 support for Fremantle, the narrow underdogs.

There's also relative uniformity in the Predictors' margin opinions as the standard deviation of their predictions is in single digits for every game in the round - only the second time that this has occurred all season.

So, which Margin Predictor has gone out on the highest limb this week?

Bookie_3 and Combo_NN_2 appear to have differentiated themselves most this week, with per game manhattan distances from virtually every other Predictor of around 10 points. For Bookie_3 this is actually a few points less differentiated than it usually is, while Combo_NN_2 is around its usual level of differentiation. Combo_NN_1 finds itself a little closer to every other Margin Predictor this week than is its wont. 

Amongst the Head-to-Head Probability Predictors, WinPred thinks the Lions deserve favouritism rather than the Tigers and that the Saints are nearer even-money chances than the TAB Sportsbet bookmaker's assessment. All three Head-to-Head Probability Predictors also think that Fremantle rather than Melbourne should be narrow favourites.

The Line Fund algorithm is particularly keen on Sydney's chances with 24.5 points start though, this week, it'll not get the chance to back that opinion with cash. It also thinks the Dons are good things giving only 8.5 start to the Roos; that opinion will have financial consequences for Investors.