It’s been pointed out to me that this week’s head-to-head wagers don’t square with the MoSHBODS probabilities in a couple of instances.

Specifically, a wager with a 33% win estimate for GWS has negative expected value at $2.80, and a wager with a 25% win estimate for Essendon is negative expected value at $3.60. These bets were, indeed, placed in error (because the order of the matches in my main spreadsheet didn’t match that in the CSVs that feed the MoSHBODS model after the Round 23 scheduling was announced). In the time-honoured MoS website tradition, those bets, because they have been made and published, will stand.

I am always cautious whenever I see the shadow-betting results of a model, because those results don’t account for the realities of actually having to get the bets on at the assumed price, and for the possibilities of errors, like mine this week in doing so. So, my practice is: if it’s wagered and published, it stands.

Now, it turns out that there was a third error in the wagering in that North Melbourne did represent value on Monday at $3.40 with a 35% win estimate, but the ship with that particular price has sailed. The best I can get now is $3.14, which still justifies a wager of 0.9% of the original Head-to-Head Fund. That wager has been placed.

So, in short, all of the wagers in the original blog stand, and we’ve added an 0.9% bet on North Melbourne head-to-head at $3.14.