2018 - Round 4 : A Few More Mismatches

As fans, we've had a very good run this season, with expected margins averaging under 15 points per game across the 27 games so far, and actual margins averaging just 27.4 points per game. (It's not hard, by the way, to show that the actual average will usually exceed the expected average - I'll do a post on this if there's any interest.)

This week, though, sees five of the games with favourites expected to win by five goals or more, which has produced an average expected margin across the nine games of almost four goals per game. That's the highest Round 4 average since 2016, and well above the figures for 2013, 2014, 2015, and 2017.

Let's have a look at what the MoS forecasters have made of it all.

TIPS AND PREDICTIONS

The first thing to note is that we're back to generally low levels of disagreement this week.

Amongst the Head-to-Head Tipsters, only Home Sweet Home (in 3 games) and the MoS twins (in 1 game, the Hawks v Dees game) have opted for anything other than the favourite in every contest.

Amongst the Margin Predictors it's once again Bookie_3, C_Marg and MoSSBODS_Marg that are lifting the overall mean absolute deviation (MAD) this week. They're all at least 4 points per game different, on average, from the all-Predictor average. Bookie_3 defines the end of the forecast ranges in five of the contests, C_Marg in three, and MoSSBODS_Marg in four. MoSHBODS_Marg is also range-defining in four of the games, but is less of an outlier when it does so, so its MAD is only 3.4 points per game.

Looking across the games, the forecasts are most disparate (in a MAD sense) in the Richmond v Brisbane Lions game where the MAD is 4.8 points per Predictor and the forecasts span a 25 point range from about a 4- to an 8-goal win for Richmond. The mean expected margin across the nine games and all Margin Predictors is 22.6 points per game, which is a little lower than the TAB average of 23.7. 

C_Prob is the Head-to-Head Probability Predictor with the highest MAD this week, even though it is the Extreme Tipster in only three contests. It's particularly dark, relatively speaking, on Richmond's and Essendon's chances. MoSSBODS_Prob has the extreme tip in four contests, and Bookie_LPSO, surprisingly, in five.

Two games have moderately high MADs, Western Bulldogs v Sydney, and Hawthorn v Melbourne, although the range of probability estimates is only 15% points and 13% points respectively in those two games, which are not especially large spans.

WAGERS

Four head-to-head and three line bets this week total about 3% of the original Combined Portfolio, which is considerably less than was risked in Rounds 1 and 3, and about the same as what was risked in Round 2.

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

Only one of the four head-to-head wagers is on a favourite, though the largest (2.4%) is on North Melbourne, who are the home team favourites. Together, the four head-to-head bets represent a little over 6% of the Head-to-Head Fund.

The largest of the line wagers is but 1.5% of the Line Fund on the Dogs, who are getting just under 3 goals start in taking on the Swans on Saturday. The two other wagers are both sized at less 0.5% of the Fund, which means that only about 2% of the Line Fund is at risk across the three wagers.

The ready reckoner for the seven wagers is below.

With less activity and smaller wagers this week, the gap between pleasure and pain is significantly smaller, a perfect set of results promising only about a 4c gain, and a worst case scenario threatening only a 3c loss. Call me old-fashioned, but a 7c swing when I'm already down by over 3c no longer fills me with dread. Maybe I've just been doing this for too long ...

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.