Football is often quite narrow in its current manifestation. Whereas in the past players would be planted across the ground, such a usage now would be considered to be a wasted resource. Wings are that for the barest of seconds before the bounce, forwards often condense up the ground. This isn’t meant to be interpreted as a good or bad thing, but merely an observation. Melbourne seem to take this to an extreme, often even forgoing the illusion of wingmen, playing as “north-south” as possible. Melbourne games often feature play that doesn’t exceed the width of the square, nor reaching over 100m in length.
This is often obscured in television coverage, but it’s sometimes more helpful to think of a modern game of football as not being played on a sprawling oval, but rather a permanently shifting soccer-sized ground within a bigger paddock. Once you win the ball from one field, you look to move it and create the next field to contest.
Corridor footy
There’s a couple of reasons Melbourne set up as narrowly as they do. Primarily, it’s their strength in the clinches that makes them able to turn marginal contested ball and stoppage situations into decisive wins. Having Max Gawn in hitout situations is like a cheat code, while Oliver, Jones, Viney, Brayshaw, et al can extract the ball from packs and get it moving like few other midfields in 2018. The narrow style increases the likelihood of these situations occurring, and provides running options for the Demons when they occur. 
The Demons are first in the league for inside 50 differential and contested ball differential this year, and they handball more often relative to their kicks than just about any other side. When the going gets tough, they get going, and the ability to explode out of tight spaces is why, when the Demons are flying, few teams have looked as dynamic in recent memory.
Of course, this comes with some trade-offs. With a condensed field, marking targets up the ground are harder to find, and the final kick inside 50 can be torture. Port were able to use this last week, with endless Melbourne forays inside 50 ending in disappointment. Out of 68 inside 50s, the Demons were only able to scrounge seven marks – one contested. Most entries were extremely predictable from the set position (mark or free kick). It was a kick to the hotspot, where Jonas, Howard, Clurey or Westhoff would kill the contest with a spoil or a mark. The Dees got just three goals from marks inside 50 (and none from inside 30).
The Power had 29 fewer inside 50s, and 16 fewer clearances, yet were able to create as many scoring shots as the Demons. That either indicates that Port’s defence was operating at historic levels, or there were fundamental issues with how the Demons went inside 50, from both a delivery and a receiving point of view.
While Port have a quality defence (and suggestions they had their worst night in two months were truly bizarre), it’s more likely an issue with the Demons. One potential suggestion as to what’s happening is that despite scoring more this year, the Dees have narrowed where they score from. It’s like they’ve been looking at Rob’s xScore charts.
Here’s a look where the Demons have scored from in the past few weeks, and how they looked against similar teams last year:
Source: Sportsmate Mobile FootyLive App
It’s is clearly visible that they have mostly narrowed their angles of attack, and taken less shots from lower value wide positions. In a vacuum, this is a great strategy! You maximise how much you score every time you get a realistic chance on goal. What could be wrong with that?
The problem is that football isn’t played in a vacuum. Against Port and Collingwood, the Demons were able to snag 9 and 12 marks inside 50 respectively. The weeks before that, against the weaker defences of the Crows, Blues and Bulldogs, they were able to snag 19, 19 and 20 MI50.
Better defensive sides can shut down effective but predictable attacks better than poor sides. There’s not much advanced statistical knowledge required for that claim. Shrinking the area teams have to defend carries the risk of blunting an attack. A strategy focused on kicking the ball to the hotspot will work best against teams who are poorer at defending the contested mark and usually-ensuing crumbs. That could be why Melbourne have been able to run up the score on poorer sides, but have struggled to get goals against better defensive units like Richmond and Port.
HPN is not suggesting that the Demons overall strategy is wrong here; just that it may need tweaking. Instead of almost completely forgoing the kicks at goal from pockets and low value wide spots, they may need to field players who can plausibly threaten from wide angles. Establishing at least the illusion that leads and shots will eventuate from there should open up other parts of the forward line.
Players like Jeff Garlett thrive in tight conditions like pockets, and the departed Jack Watts converted at a high rate shooting from the easier set shots available via marking near the boundary lines. They both contributed to those wider angled shots in the 2017 shot charts we examined, and obviously while Watts is gone, Garlett is languishing in the VFL and could be used right now.
Balance matters, especially when attempting to confuse defences that have ever-increasing footage and intelligence on how you play.
Critical backline issues
We mentioned yesterday in our mid season review that Melbourne’s biggest struggles have come down back, and their inability at times to stop opponents from scoring. Like the problems in attack, these mostly materialise when facing good teams, and not bad ones. The Demons are conceding points per inside 50 like a bottom six side, at a rate that no recent Grand Finalist has conceded.
With the recruitment of the now-missing Jake Lever, the Demons had three defensive members of the 2017 All Australian squad in their best 22 this year, with no other side having more than two (West Coast, and that is sticking with Yeo as a defender, which he may not be). On paper, they strengthened their biggest perceived weakness over the offseason, and added an extremely talented young player who should have surely meshed with their relatively successful current setup.
Except it hasn’t remotely worked like that.
We briefly mentioned “the fit” of Lever around other Melbourne talls being a potential issue when the Lever trade was first made, without going into a lot of detail. What has eventuated is that Tom McDonald headed forward, and Sam Frost headed to Casey, which opened up a role for Lever as the second/third key position defender and intercept marker – a specialised role in today’s game. The problem with that is that Melbourne already had one of those types in 2017 All Australian Michael Hibberd. He is 6cm shorter than Lever but he also works best when slightly less than totally accountable. Having one of these instinctively aggressive intercepting markers is fantastic for a side, but two might cause other problems.
Against practically every team, having at least two accountable tall defenders and a player adept at sliding off their man to intercept should be critical. Melbourne has one true tall defender (the ever-improving Oscar McDonald) and a bunch of options masquerading against type as the second and third defender options. Jake Lever, Michael Hibberd, Joel Smith, Sam Frost and Neville Jetta have all been shuffled through those roles, without finding a good balance. This week, Harrison Petty gets a shot. Nothing has worked to a level that Melbourne would be happy with yet.
This has had a noticeable effect on all three AA squad members from last year. Their Player Approximate Value (PAV) numbers are down in gross and marginal per game terms, and they have declined on the AFL’s Player Ratings. Looking more discretely at the numbers, one in particular has struggled this year, despite calls for him to be in the All Australian team.
Neville Jetta has undertaken a complicated journey from being an ignored lesser Jetta, to underrated, to correctly rated to being the best Jetta and perhaps a bit overrated. The last step is a recent occurrence, and perhaps a sign that it is the hardest to rate defenders of all players. PAV wasn’t in love with Jetta last year (rating him as by far the worst player in the AA squad), but all rating systems have blind spots (minor and major). What was interesting is that the other major system, AFL Player Ratings, also didn’t rate him in the best dozen small/mid defenders.
It seems that Jetta was passing many an eye test, but failing the data one.
We decided to have a look at some more advanced information about Jetta, and follow his game closely in person last week at Adelaide Oval after attending the Footballistics launch.
Watching Jetta live, there is no doubt that he is a courageous and energetic player. He never shirks a contest, and seems to put his own safety second to that of his team’s needs. He is surely among the hardest working players in the league, often covering for his teammates’ poor decisions and providing a contest. But he also loses a lot of contests himself, and can place himself in bad spots to stop the opposition.
Watch this highlight package of Port’s 11 goals last week.
Two goals are created from free kicks that Jetta conceded. A further two came from marks where Jetta was the closest defender to spoil or stop. And one more came when Jetta made a late decision to charge the ball carrier, who then passed to his direct opposition near goal (which was the right move by Jetta, but worth noting for this point nonetheless). That’s about four goals of eleven that could be categorised as scored against Jetta, more than any other Melbourne defender. And he still got mentioned as one of the best players on the ground on the coaches votes.
Not all of these were Jetta’s fault – on at least three of the five occasions above he was somewhat covering for the mistakes of others. But cleanup should not necessarily be his role.
Using numbers sourced from AFL Pro Stats, so far this year, Jetta sits third last for all defenders in Contested Defence One on Ones (CDOOO) loss % (minimum 6 games played, 2 CDOOO occasions per game) out of the 53 defenders who qualify. For small/mid defenders, he is last using the same qualifiers. CDOOO’s may not be the clearest stat in the world (HPN believes they don’t include direct one-on-one lead situations, amongst others), but it is probably still indicative, especially comparing the same player to his past performance. This year Jetta has lost 39% of such contests, last year his rate was 27.5%. There has been a slip.
By contrast, Jake Kolodjashnij has lost just 21% of these contests this year, Jeremy Howe just 16%, Heath Shaw 15%, Jake Kelly 15%. Jetta also sits behind a large number of players in Defensive Half Pressure Acts and Intercept Marks. With the other quality ball users in the Dees back half, he also isn’t relied upon to rebound the ball back outside 50.
Instead, he often has to take the opposition’s second most dangerous tall forward, a tough ask for a player who is 180cm tall. Jetta is very crafty and has some agility, but this is perhaps the worst possible way to use him.
Harrison Petty is the option that Melbourne have tabbed to fix the problem this week, and if it works the Dees should be able to redeploy Jetta more effectively. If it doesn’t, the options aren’t looking great. One option the Dees haven’t explored this year is sending Tom McDonald back to defence, and sacrificing his forward impact. This would be a drastic move, but McDonald was one of the finest defenders in the game before he was shifted around the ground last year. PAV rated him as the second best defender behind Rance in 2014 and 2015, though he received little in the way of plaudits for his efforts. The dilemma is that McDonald is probably in the top six or seven forwards in the competition right now, and the brains trust would cop a lot of flak for moving such a productive player away unless it visibly paid off quickly.
There will have to be the discussion at some point whether a Weiderman/Hogan combination, with Hogan staying at home more like in previous years, will be able to provide a slightly worse version of what McDonald/Hogan do now. It may be worth it if McDonald is able to drag the Demons defence to the sort of league average level they probably need to contend.
These are conversations best had at the start of the year and not in the middle, but the Demons are running out of time.