The Weekend Preview is a blog that aims to provide a different angle on the narrative of an upcoming sporting event.

AFL Round 12: Melbourne vs. Collingwood

AFL Round 12: Melbourne vs. Collingwood

Monday Afternoon, 11th June 2018 at the MCG 

----------

It’s probably the No. 1 versus No. 2 ruckman in the comp. It’s an exciting battle. The ruckmen do have a legitimate contest. Forwards these days rotate so heavily through the midfield, defenders pick up certain guys depending on where they go. Brodie versus Gawny will be a legitimate fight for that crown so to speak.

Scott Pendlebury [1]

----------

The ruck battle is the last great head-to-head duel in modern football. And this week the two premier ruckmen in the AFL this season, Max Gawn and Brodie Grundy, will meet for battle at the MCG on Queen's Birthday afternoon. But which one can lay claim to be the best ruckman in the league?

Table-1 presents some summary season statistics for both players. At a high level, Gawn would appear to have the upper hand in the ruck contest and Grundy when the ball is in open play. But which player has had the greater influence on match results?

 Table-1: Max Gawn and Brodie Grundy Comparison

Table-1: Max Gawn and Brodie Grundy Comparison

As most teams play only the one specialist ruckman, we can quantify a ruckman's contribution to a match not by their overall individual statistics, but their relative performance against their direct ruck opponent.

But before we can do this assessment, we first need to determine the general relationship between winning a given individual statistic and the match result. To determine this, we shall select a representative performance statistic from the following 4 categories of play:

  • Ruck Contest: Hitouts-To-Advantage.
  • Open Play: Contested Possessions.
  • Marking Contest: Contested Marks.
  • Direct Scoreboard Impact: Goals.

[Note that we have purposefully limited the assessment to only four variables in order to maintain some mutual exclusivity between statistics. In reality there is crossover, for instance a contested mark is a special case of contested possession and a goal may occur directly from a contested possession or a contested mark.]

We shall consider each of these statistics in turn and determine how much they contribute to an overall match result. Melbourne and Collingwood shall be used as case examples.

Hit-outs to Advantage:

A Hitout-To-Advantage (HOA) is defined as follows:

  • Hitout-To-Advantage (HOA): A hitout that reaches an intended teammate. [2]

A HOA is more than a simple hitout; it is a hitout that purposefully and directly contributes to the next chain of play. The relative importance of the HOA is obviously dependent on which part of the ground it occurred. And of course some teams that are lacking a quality tap ruckman may purposefully concede the ruck battle and setup at the stoppages accordingly, nullifying the possible positive impact the HOA may otherwise have had. So is winning the hitouts and specifically the HOA an irrelevance?

Figure-1 presents the relationship between Match Margin and Hitouts-To-Advantage differential for both Melbourne and Collingwood in season 2018. Melbourne and Collingwood are ranked 1st and 2nd in the AFL for HOA. However, Melbourne has a positive correlation between HOA and match margin, whilst Collingwood has a negative correlation (identified by the slope of the trendline). This all demonstrates the difficulty in converting ruck domination to scoreboard impact. This is also highlighted when considering the AFL as a whole and the varying approaches towards the ruck contest. Figure-2 presents the relationship between Match Margin and HOA for all matches played in season 2018. As a whole there is a small positive correlation, with the slope of the trend line providing an indication of the overall relative benefit of winning the HOA differential. This can be summarised as follows:

  • +1 Hitout-To-Advantage = 0.73 points
 Figure-1: Melbourne & Collingwood – Match Margin vs. Hitouts-To-Advantage Differential

Figure-1: Melbourne & Collingwood – Match Margin vs. Hitouts-To-Advantage Differential

 Figure-2: All 2018 Matches – Match Margin vs. Hitouts-To-Advantage Differential

Figure-2: All 2018 Matches – Match Margin vs. Hitouts-To-Advantage Differential

Contested Possessions:

A Contested Possession is defined as follows:

  • Contested Possession (CP): A possession which has been won when the ball is in dispute. Includes looseball-gets, hardball-gets, contested marks, gathers from a hitout and frees for. [2]

The contested possession differential, i.e. how many contested possessions a team accumulates relative to their opponent, is the common go-to Key Performance Indicator (KPI) for a team’s performance. It has historically proven to be a good indicator of performance and season 2018 is no different.

Figure-3 presents the relationship between Match Margin and Contested Possession (CP) differential for both Melbourne and Collingwood in season 2018. Both teams are strong contested ball teams (ranked 1st and 2nd), and in the case of Melbourne they haven’t lost the differential all year. But critically, for both teams there is a positive correlation between contested possession differential and match margin. That is, the more contested possessions the team wins relative to their opponent, the larger the expected match margin, and as result the more likely they are to win the game. And this is consistent with the league as a whole. Figure-4 presents the relationship between Match Margin and Contested Possession differential for all matches played in season 2018. The positive correlation with match margin is far more noticeable for contested possessions than for HOA. The overall relative benefit of the CP differential is approximately double that of the HOA, and can be summarised as follows:

  • +1 Contested Possession = 1.34 points
 Figure-3: Melbourne & Collingwood – Match Margin vs. Contested Possessions Differential

Figure-3: Melbourne & Collingwood – Match Margin vs. Contested Possessions Differential

 Figure-4: All 2018 Matches – Match Margin vs. Contested Possessions Differential

Figure-4: All 2018 Matches – Match Margin vs. Contested Possessions Differential

Contested Marks:

A Contested Mark is defined as follows:

  • Contested Mark (CM): When a player takes a mark under physical pressure of an opponent or in a pack. [2]

The contested mark is a subset of the contested possession but its relative importance is worth highlighting. The ability to take a contested mark can be an invaluable asset to a team, whether it is the goal saving mark in defence, the relieving down-the-line pack mark on the wing, or the forward-50 one-on-one.

Figure-5 presents the relationship between Match Margin and Contested Marks differential for both Melbourne and Collingwood in season 2018. Similar to contested possessions, both Melbourne (ranked 1st) and Collingwood (ranked 5th) are strong contested marking teams. And the positive correlation between contested marks differential and match margin is once again evident. Figure-6 presents the relationship between Match Margin and Contested Possession for all matches played in season 2018. The overall relative benefit of the Contested Mark is approximately double that of a general Contested Possession and hence is approximately 4 times as important as a HOA:

  • +1 CM = 2.87 points
 Figure-5: Melbourne & Collingwood – Match Margin vs. Contested Marks Differential

Figure-5: Melbourne & Collingwood – Match Margin vs. Contested Marks Differential

 Figure-6: All 2018 Matches – Match Margin vs. Contested Marks Differential

Figure-6: All 2018 Matches – Match Margin vs. Contested Marks Differential

Ruck Rankings:

As was determined in the previous sections, we now have methods for converting differential for HOAs, CPs and CMs into a direct scoreboard impact. And if you then include goals, we can summarise as follows:

  • +1 HOA = 0.73 points
  • +1 CP = 1.34 points
  • +1 CM = 2.87. points
  • +1 Goal = 6 points

We can now use these weightings to determine the estimated net impact on a match that a ruckman has. To do that we need to assess the head-to-head battle of the primary ruckman in every game this season and determine the net scoreboard impact they are having in each game based on their differential performance in the above statistics.

The result of such an assessment is presented in Figure-7. The figure presents the estimated net scoreboard impact of the leading ruckman in the AFL this season, i.e. on average how much they are influencing the match result. And from this summary, we can conclude that Max Gawn has indeed been the most effective ruckman in the AFL this season, averaging an estimated net positive impact of just over 3 goals (18.5 points) per game. Second in the rankings is Brodie Grundy who is averaging an estimated net positive impact of 13.3 points per game.

 Figure-7: Round 11 Ruck Rankings

Figure-7: Round 11 Ruck Rankings

Note: This article represents a continuation of a series of articles to determine the form players of the competition in each position. The premise is to assess each position based on an easily identifiable collection of variables. Each week the rankings will be updated and an All-Australian team will be auto-populated and presented on a stand-alone page. So far, metrics for rucks, key forwards, wingers, attacking halfbacks and key defenders have been developed.

Match Tip:

In all the metrics discussed in this article, Melbourne are ranked number 1 in the AFL. Collingwood are ranked second in Hitouts-To-Advantage and Contested Possessions.

 Table-2: Melbourne and Collingwood Performance Metrics Summary

Table-2: Melbourne and Collingwood Performance Metrics Summary

On current form, Collingwood are worthy of a spot in the Finals this year. However, it is hard to deny the form and, dare I say it, the premiership credentials of the Melbourne Football Club.

Melbourne to win by 26 points

----------

    [1] Cherny, D. (2018).“Gawn, Grundy reaching for ruck crown: Pendlebury”. The Age, 7th June 2018.

    [2] Champion Data. (2018). “2018 AFL Prospectus”. Champion Data Pty Ltd.

    Note: Data used in this article was sourced from AFL StatsPro on the afl.com.au website.

    FIFA World Cup: France vs. Australia

    FIFA World Cup: France vs. Australia

    AFL Round 11: Sydney vs. Carlton

    AFL Round 11: Sydney vs. Carlton