WoSHBODS tipped 8 from 9 winners this week and the bookmakers all 9, and so is now 5 tips behind them both across the whole season.

Its Margin MAE was 17.2 points per game, compared to 19.1 and 19.8 for the two bookmakers, and its Total MAE was 22.9 points per game, compared to 24.9 for the two bookmakers.

WoSHBODS’ season-long Margin MAE is now 19.1 points per game, compared to 18.5 for the two bookmakers.

Lastly, WoSHBODS’ LPS of +3.5 bits compares to +5.6 bits for the two bookmakers.

Across the season WoSHBODS is on +26.7 bits to +33.5 and +33.4 for the two bookmakers.

WAGERS

The Line and Overs/Unders Funds made money this week and more than offset the losses by the Head-to-Head Fund such that the value of the Combined Portfolio rose by2.5c on the round leaving it now down by just 1.6c on the season, that from a -1% ROI on a 1.6 turn.

That’s quite the recovery from being down by over 9c after that horror Round 6.

TEAM MARGIN MAEs

Let’s have a final look for the season at how WoSHBODS has performed on Margin MAEs on a team-by-team basis.

It’s done best on Sydney, GWS, and Essendon games where it has enjoyed margin MAEs of around 14 or less. It’s also gone under three goals a game in MAE for six other teams.

Games involving Geelong especially, but also Hawthorn and Brisbane have been most problematic, with margin MAEs for those games ranging from about 23 to 30 points per game.

Ten teams have margin MAEs below the average of 19.1 points per game, and eight teams have margin MAEs above the average.

TEAM LPS

Next, the team-by-team LPS’ to get a feeling for how accurately WoSHBODS has been assigning probabilities to each team’s chances.

Here we find WoSHBODS doing best with St Kilda and North Melbourne, and worst with Hawthorn, Gold Coast, Brisbane and West Coast.

Comparing these results to those for Team Margin MAEs we find that:

Richmond, Essendon, and Collingwood are in the Top 6 on both metrics
West Coast, Brisbane, Gold Coast, and Hawthorn are in the Bottom 6 on both metrics
Geelong is in the Top 6 on LPS but is last on Margin MAE

Only five teams have contributed negative bits when summed across all 12 rounds.

PERFORMANCE BY MARGIN

As one last thing to look at for the last time this week, let’s see how WoSHBODS has performed on depending on the final margin of the game.

We see that the smallest MAEs have come in the games with the narrowest victory margins - those won by less than three goals - with the games most inflating the overall MAE the 19 where the margin has been 8 goals or more.

Conversely, accuracy has been greatest for games with the largest margin and near coin-toss level in games decided by less than a goal.

TEAM DASHBOARD

Below is a table recording how the teams rank on various Team Dashboard metrics.

The metrics most rank correlated with ladder position are:

MoS Win Production Function: +0.97
Percentage of Quarters Won: +0.95
Own Goals per Game: +0.90
Own Points per Game: +0.89
Own Scoring Shots per Game: +0.88
Opponent Goals per Game: +0.86
Opponent Points per Game: +0.85
Q4 Performances: +0.85
Q2 Performances: +0.83
Opponent Scoring Shots per Game: +0.81
Q1 Performances: +0.80

The metric least rank correlated with ladder position is:

Opponent Scoring Shot Conversion: +0.42

Note that the teams in the Top 8 on the ladder and also in the Top 8, albeit in a different order, on the Win Production Function.

Sydney rankings include bottom half finishes on Opponent Scoring Shots, Opponent Goals and Points, and Q1 and Q2 Performances, while Hawthorn rankings do likewise on Opponent Scoring Shot Conversion, Opponent Goals and Points, and Q1 and Q2 Performances.

Next come the details of the Team Dashboard in two parts.

In the top part we can see that the Top 9 teams represent 74% of all wins.

We can also see that Geelong has scored an average of almost 11 goals a game and conceded under 6 goals, while Melbourne and North Melbourne have scored about 9 but conceded only about 3-and-a-half to four-and-a-half. St Kilda has scored just under 4 goals per game and conceded over eight.

Looking at the WIn Production Function:

Sydney, Gold Coast, and Port Adelaide have all won about 1-and-a-half games more than expected, based on their scoring data
Carlton, Geelong, Richmond, and Western Bulldogs have won about 1 game more than expected
GWS and Brisbane have won about 2 games fewer than expected
Sydney, Fremantle, and Collingwood have won about 1 game fewer than expected

Also, Collingwood have converted at over 15% points lower than their opponents (36% v 51.4%) while Melbourne have converted at over 15% points higher than their opponents (53.3% v 37.9%).

In the bottom half we see that:

North Melbourne has won 38 (and drawn 1) of 48 quarters this season. Geelong is next with 36, then Melbourne with 32 wins and 2 draws and then Gold Coast and Carlton each with 28 wins and 2 draws.
Over the same time period, Collingwood has won only 12 quarters and drawn 1, while St Kilda won 12 and drew 2. Essendon finished 14 and 1 draw, and Richmond 14 and 2 draws.
The teams that have outscored their opponents, on average, in every quarter are Geelong, North Melbourne, and Melbourne. Gold Coast have done the same for Q2 to Q4 also, but have an 88 percentage for Q1. Western Bulldogs have done the same for Q1, Q2 and Q4, but have a 51 percentage for Q2. Hawthorn have done the same for Q2 to Q4, but have a 59 percentage for Q1.
Those that have been outscored by their opponents, on average, in every quarter are Collingwood, Essendon, Richmond, St Kilda, and West Coast. Brisbane have done the same for Q2 to Q4 but have a 139 percentage for Q1. Gold Coast have done the same for Q2 to Q4 also, but have a 139 percentage for Q1. GWS have done the same for Q1, Q2 and Q4, but have a 118 percentage for Q3.
Western Bulldogs have scored only 15% of their points in Q3s.
GWS have scored 35% of their points in Q3s.
Carlton have conceded only 14% of their points in Q1s.
Hawthorn have conceded 36% of their points in Q1s, Carlton 37% in Q2s, and Collingwood 36% in Q3s.
Collingwood have converted a tonly 26% in Q2s.