2016 - Round 10 Results - That's My Line

This week saw a return to profitability - the first in four weeks - thanks solely to a season-high +7.6c return for the Line Fund from its 2 from 2 performance. The Head-to-Head Fund missed with its only two attempts, meaning that it has now shed almost 25% of its initial value over the last six rounds. Nonetheless, thanks to its remarkable start to the season, it remains up by 10.5c overall.

The Overs/Unders Fund managed just 3 from 6 for the weekend, shedding 0.6c to leave it down 1.4c on the season. 

In aggregate, the Overall Portfolio rose by just under 1c to finish Round 10 up just over 22c for the year-to-date.


On the whole, it was a good weekend for the ChiPS family of forecasters and a relatively poor, but not dire, one for the MoSSBODS family.

In the Head-to-Head Tipping competition MoSSBODS_Marg's score of 7 from 9 was slightly above the all-Tipster average of 6.89, however, and was enough to preserve its two-tip lead over BKB at the top of the MoS Leaderboard.

C_Marg's 8 from 9 was equal-best for the round, matching the score of both RSMP Tipsters, and leaving C_Marg just one tip adrift of BKB. Home Sweet Home and Consult The Ladder had the round's smallest hauls of just 5 from 9.

For the Margin Predictors, Bookie_LPSO recorded the lowest mean absolute error (MAE) of 17.4 points per game, which propelled it into 2nd-place behind C_Marg, whose 20.0 MAE saw it move back into 1st-place for the first time since Round 3. The all-Predictor average MAE was 21.1 points per game per Predictor.

MoSSBODS_Marg returned a relatively disappointing 23.9 MAE and slid into 6th-place on the MoS Leaderboard, the lowest position it's held since Round 3.

Bookie_Hcap remains in last place, a reflection of just how poorly the opening lines on the TAB have foreshadowed final margins this season. Long may that continue ...

Amongst the Head-to-Head Probability Predictors, Bookie_LPSO returned the highest probability score, doing slightly better than the two other bookmaker-based Predictors and C_Prob.

MoSSBODS_Prob paid heavily for its assessment of the Blues as only 7% chances against the Cats while the bookmaker-based Predictors had them as 14-16% chances and C_Prob had them as 10% chances. It's in the nature of the logarithmic probability scoring used in MoS that "bolder" but ultimately incorrect assessments are penalised, all the moreso the closer they are to 0% for some event that actually occurs (see the chart at the bottom of this blog post).

That result for MoSSBODS_Prob has shunted it to the bottom of the pile of Predictors, while C_Prob's performance has left it in 1st-place - for now. The bookie-posse is circling ...