Matter of Stats

View Original

2018 - Round 11 : Low Scores, Big Margins

We return to a full menu of nine games this weekend though a few of them look fairly unappetising, as conveyed by the all-game average expected margin this week, which has come in at 28.7 points per game. That's the highest single round average in the home and away season since Round 23 of 2016. 

What's driving up the average this week are expected margins of more than four goals in six of the games, and expected margins of over 20 points in two more.

It's only the Hawthorn v Port Adelaide game where it seems fans can have any real hopes of a close finish.

Fans have, in fact enjoyed relatively few close games this season, which seems odd in the context of the low levels of scoring we've all been reading out. The following graphs give some historical context to the season's scoring.

What we see is that, relative to other seasons in the modern era in particular, 2018 has produced lower totals, fewer blowouts, but fewer close games as well. Overall though, average victory margins have been around about the era average.

To the MoS forecasters.

TIPS AND PREDICTIONS

With so many firm favourites this weekend, it's to be expected that disagreement levels will be low for the Head-to-Head Tipsters, and this is indeed the case. In fact, the only thing preventing a first-ever zero Disagreement Index is the fact that there are four away team favourites. Enter Home Sweet Home ...

So, no meaningful movement on the Leaderboard this week then.

Amongst the Margin Predictors, it's MoSSBODS_Marg's turn to have the largest mean absolute deviation (MAD), at 7.4 points per game. MoSHBODS_Marg has the second-highest MAD, and Bookie_3 third-highest, relegating C_Marg to an unfamiliar fourth placing on this metric.

Relative to the current leader, Bookie_Hcap, a best set of results for MoSSBODS_Marg would see it move from about 97 to about 19 points behind, while a best set of results for MoSHBODS_Marg would see it move from about 100 to about 29 points behind. That's incredibly unlikely to happen, but relatively small wins by Melbourne and West Coast, and large wins by Port Adelaide and Richmond, would help close those gap at least some of the way.

At the game level, there are two matchups with MADs of over 6.0 points per Predictor, West Coast v St Kilda, and Sydney v Carlton, with the MoS twins and C_Marg significant contributors to the MADs for both of those games. Overall, disagreement levels are moderate.

The mean expected margin across the nine games this week and all Margin Predictors is 28.6 points per game. 

Turning lastly to the Head-to-Head Probability Predictors we find C_Prob as the Predictor most extreme in six contests, Bookie_LPSO in four, and MoSSBODS_Prob in three. MoSSBODS_Prob's particularly low probability estimate for Adelaide, however, sees it with the round's highest MAD of 4.7% points per game.

MoSSBODS_Prob's low probability estimate has also contributed to the fact that the Adelaide v GWS game carries this week's largest MAD of 5.4% points per Predictor, though there are three more games above 4.5%. Overall disagreement levels are low, however.

WAGERS

Investors face six head-to-head and five line wagers this week (though the algorithm did recommend a $0 line wager on the Dogs, which I failed to get any bookmaker to accept). The largest of the head-to-head bets is just under 4% on Port Adelaide at $1.90 in the coin-toss with the Hawks, and the largest of the line bets 3.0% on GWS +22.5, also at $1.90

(Please click on the image below to access a larger version.)

The ready reckoner, which quantifies those wagers' individual and collective risk, appears below, and shows a less even spread of risk than Investors have faced in recent weeks.

Five games carry potential swings of over 2%, greatest amongst them the possible 4% swing in the West Coast v St Kilda game, where the maximum upside is +2.8% and the maximum downside -1.2%. There's also a 3.1% swing in the Hawthorn v Port Adelaide game, a 2.9% swing in the Adelaide v GWS game, a 2.6% swing in the Essendon v Richmond game, and a 2.1% swing in the Gold Coast v Geelong game.

Other than that, there are no bets in two games, and more modest potential swings in two more.

In total, almost 7% of the original Combined Portfolio is at risk and the maximum upside is almost 10%.

To finish, here are MoSHBODS' and MoSSBODS' opinions about likely team scoring.

Commentary about these forecasts will come in the week's Overs/Unders blog.