2011 Round 15: No Attachment

Continuing Sunday's theme, this week I'm choosing to be unattached to the results that the round's football provides (we'll see how long that lasts ... I give it until about 8pm Friday night).

This weekend there's less to be attached to anyway since we've no bet in three of the eight games and only a perfunctory 0.4% line bet in another. All told, we've just 2 head-to-head bets totalling just under 10% of the Head-to-Head Fund, and 5 line bets totalling almost 15% of the Line Fund.

Financially, the game that matters most is Sunday's Pies v Hawthorn clash where a 23 points or greater victory by the Pies will add just over 4c to the Portfolio value, and an upset Hawks win will knock over 7c off that same value.

A Tigers win promises the next largest gain, about 2.6%, while a win by the Blues instead in that game will cost Investors 1.4%. Other significant results, good and bad, are a Fremantle win by 40 points or more, worth 1.9%, the flipside of which is a 2.1% loss should any other result eventuate, and a Roos win, draw or loss by 1 or 2 points, which is worth 1.3% and which would prevent a 1.4% loss.

For all but one of this week's games at least 10 of the 13 Head-to-Head Tipsters agree on the winners. The Dogs v Melbourne game is the exception where these Tipsters are split 7-6 in favour of the Dees. In those games where there is strong majority support it's with the favourites apart from in the St Kilda v Kangaroos contest where the Tipsters are behind the underdog Roos 10-3.

Majority opinion for the Margin Predictors matches that for the Head-to-Head Tipsters except for the Saints v Roos game where the Margin Predictors are favouring the Saints by 10-3. That said, none of them is tipping a runaway Saints victory: the largest predicted victory margin is 11 points by Combo_NN_2 and the next largest is just 6 points.

Generally there's not much divergence of opinion amongst the Margin Predictors, the standard deviation of the predicted margins being 10.4 points or less for every game.

There's also not a lot of difference in Predictors' margin predictions this week when you consider how different they are from the all-Predictor average in each game compared to how different they normally are. (I'm sure there's a better way to phrase this but right now it's eluding me.)

Only Combo_NN_2 can make any claim to being differently different this week. Across the season it's tended to favour Away teams by 0.1 standard deviations more than the average Predictor but this week it's tending to favour Home teams by 0.8 standard deviations more than the average Predictor. This is the second time in three weeks that it's done this.

Just the Probability Predictors then before we're finished. ProPred and H2H are in complete agreement with the TAB Sportsbet bookmaker about which team should be the favourite in each game and, roughly, what that favourite's chances of victory are. WinPred disagrees with the trio's choice of favourite in two games, opting to make the Dees slight favourites over the Dogs, and the Roos favourites over the Saints.

The Line Fund algorithm rates four teams as 65% or better chances of winning on line betting. It has Geelong as 69% chances of covering the 44.5 points spread they're offering the Dons, Collingwood as 68% chances of covering the 22.5 points spread they're offering the Hawks, Port Adelaide as 66% chances of preventing the Lions from covering the 18.5 points they're offering, and Sydney as 65% chances of covering the 15.5 points they're offering the Crows. Three of these four teams are Away teams, which is why you're not (now) seeing Line Fund bets for these bets, despite the high level of confidence.