Before the year started, HPN predicted that Adelaide and Port Adelaide were likely to be the best two sides in the competition. Things have not exactly gone to plan for either side so far this year; both sides have been hit by injuries, and other issues around incoming/outgoing talent. Both are starting see their selected sides strengthen, and they could yet hit those predicted heights; helped by a good showing this week, that is.
With hindsight the Showdown could still turn out to have featured two of the best sides at year’s end, so let’s look at some of each side’s issues and strengths so far to see how they will potentially get to those heights.
Port’s defence should get better
Last year, Port succeeded by stopping opponent scoring, rather than scoring a lot themselves. A stout team-driven approach denied opponents opportunities near goal, with the Power trailing only Sydney and Richmond in allowing the fewest points per inside 50 entry conceded (as well as goals and scoring shots). They denied set shots at goal, the side who allowed the fewest marks inside 50 per opposition entry.
Marks inside 50 per inside 50 (MIFPIF) correlates with quality defence over the last 20 years or so. The better a team is at denying marks in dangerous spots, the better they tend to be in suppressing scoring shots, goals, and points per inside 50. We’ve charted the scoring shots correlation below:
But the converse measure, MIFPIF obtained in attack, isn’t as effective at predicting offensive success (we will get to that a bit later):
This year the Port Adelaide tall defence is holding up, allowing a miserly MIFPIF of 16%. Only Geelong has done better, with 15% so far this year. By contrast the Bulldogs allow marks at a rate of 29% of opponent entries, and are also the fourth worst defensive team this year on the measures discussed above.
According to the linear trend line above, a side conceding 16% of MIFPIF should only concede scoring shots on 33% of their conceded inside 50s. Instead, Port Adelaide are conceding on a massive 42% of inside 50s – a major outlier, and one without a clear discernible reason. The Power are conceding both more scoring shots and goals per i50 entry than they did last year – which isn’t the way it usually works when the other indicators are so positive.
Strikingly, while league-wide accuracy is down, teams are more accurate against Port Adelaide than they were last year. The shot charts for their previous games don’t really indicate why, beyond some sides (Brisbane, Geelong, West Coast) being relatively accurate against them from the boundary and outside 50 :
(screenshots taken from the FootyLive app)
Just looking at Port during gameplay, visually their approach seems similar to last year. Tom Jonas is in borderline All Australian form, and kills a lot of contests in the air. Dougal Howard’s extremely hot spoiling start has cooled a little, but he is playing a role at a standard not expected by many outside of Alberton. While some bit players are being relied on for important roles (Barry, Bryne-Jones, the corpse of Trent “Superboot” McKenzie), the structure seems to be holding up, and for the most part keeping opponents from the high value parts of the 50.
Given the current discrepancy between the measures is unusual, one of two things could happen from here. One possibility is that the MIFPIF is a trailing indicator, and the Power will start to slide in that measure as well, conceding more marks to move in line with the actual scores being conceded. Alternatively, the Power will be a little “luckier”, as shots like the more speculative goals of previous weeks become behinds in weeks to come.
If we had to bet on these options, our money is firmly on the latter, just based on the historical data and the “eye test”. There doesn’t appear to be any critical flaws with the backline setup, and it resembles the dominant model of last season. Unsustainable opponent accuracy should revert back to league average, and the Power backline should be one to feared again.
Despite injuries, Adelaide’s forward line is still straight fire
Port’s defensive bounce, if it happens this week, will have to come at the expense of the most consistently excellent forward unit in the league across the past three years. And it might very well be the battle that the match turns on.
If you haven’t been living under a rock, you will have heard that scoring across the league is down.
Which is true!
Last year, teams scored about 1.70 points per inside 50, and this year it is down to about 1.54.
Adelaide doesn’t care about the rest of the league. They just keep on scoring and scoring. This year, they are getting 1.82 points per inside 50, down marginally on their 2017 output of 1.87, but still faring much better than the rest of the league.
Right now, they score 0.28 points extra than the average team per inside 50 – which doesn’t sound much, but over the course of a game (and a typical 52 inside 50s per team) that’s about 15 points more than an average teams attack.
While it may feel a little too mechanical to break down football like that, it is a good way to describe the effect of the Adelaide forward line. They are inevitable, and the best way to stop them is to stop the ball getting near them.
That average team would need a 9-10 inside 50s advantage to shift the deck back to level – which is a fair margin. In the Grand Final last year, Richmond did just that, winning the inside 50 count 58 to 49. Richmond were also powerful down back, with a counter to nearly all of Adelaide’s potent weapons – but the point remains.
This year’s forward dominance has occurred despite Adelaide losing the previously underrated Charlie Cameron in the offseason, who is currently tracking as the best forward in an admittedly struggling Brisbane offensive lineup. They’ve also had Betts, Lynch and Walker unavailable for chunks of the season to date. Last year, the Crows dominated in the air inside forward 50, leading the league for MIFPIF (marking on 25.5% of all entries, give or take entries with multiple marks). This year, and perhaps as a testament to the changing nature of their forward line, they are down to just 19%, which is roughly league average.
This is somewhat in line with the point above, notably that the correlation between marks, scoring shots and points per inside 50 on the defensive side of the ledger doesn’t quite carry over as strongly to attack.
Good teams can find a variety of ways to score, and Adelaide have done so. Like last year, the attack doesn’t have a singular focal point, which adds to the difficulty of stopping them. Where some sides have hegemonic targets scoring more than 20% of their total score, such as Brown and Franklin, the Crows haven’t had anyone score more than 14% of their total score across either of the last two seasons. This season Gibbs, Douglas, Atkins and Seedsman have all hit the scoreboard more than Betts and Lynch this year, which isn’t going to last, but is still fun for now. And it shows that everyone is a threat.
The mere fact that the Crows have such a deep forward line allows them to cover for an absence (or two) much more easily. When an All Australian level forward like Taylor Walker is out, the rest of the forward line is versatile enough to shift up, so the replacement doesn’t have to be the number one (or two) option. Not only are the Crows forwards talented, but they also have a diverse skillset; from solid contested marks (Walker, Lynch, Jenkins, McGovern) to roaming, leading marks who play can play up the ground (Walker, Lynch and to a lesser degree McGovern), to extremely dangerous small forwards who always need watching (Betts and Milera). Defences can’t really hide weak performers down back against them, and conversely the Crows can afford to rest a mid (or a lesser player) up forward without really damaging the potency of their attack.
Their marking inside forward 50 should bounce up a little as the season progresses, especially with players such as Walker and Lynch play more footy. Regardless, it sets up a good battle against what should be a resurgent Port defence.
Port’s free agents haven’t really fit in yet
Of Port Adelaide’s crop of much vaunted recruits in 2018, none have quite lived up to expectation so far. Motlop perhaps has come the closest, but is still down in most counting stats on his past four years. Jack Watts has looked more like the overmatched 2017 version rather than the fully-realised, super second forward of 2016. Rockliff has been hampered by injuries, Barry has fit into a minimal role at the bottom end of the selected 22 (before being dropped this week), and McKenzie has made a brief cameo. Perhaps the most important player involved in Port’s 2017 trade period so far this year has been Jarman Impey, who has impressed in a well defined role at Hawthorn.
That defined role (or roles) could be some of the issue – namely the fit within the Port Adelaide system.
Let’s start with Jack Watts. He thrived at Melbourne when playing as a higher leading forward who converted from range but had the speed to lead back deeper to goal with Hogan fit. He struggled significantly when he was required to be the focal point of the Demons attack, or even worse the ruck option. With Ryder’s injury, Watts has become the main focus. Port were forced to push Charlie Dixon (amongst others) into the ruck – hurting both his and Watts’ forward impact. To his credit, Watts is averaging a career high tackles per game so far this year, quietening some of the criticism about his “toughness” (whatever that is). At least they didn’t do what Melbourne tried, and use Watts as an emergency ruck (it’s hopefully in his contract to avoid that job at all costs).
Having to spend time rucking, Dixon is scoring roughly half as many points per disposal as he did last year. This would work if others could pick up the slack, but from Watts and from others, that just hasn’t happened yet. Todd Marshall looked promising until injury and personal issues took him off the field, and the rest of a somewhat makeshift forward line is still trying to find their place.
Additionally, with Powell-Pepper’s suspension and Rockliff’s injury, Port have been forced to push Robbie Grey into the clearance battle more than they did in 2017, which again throws the likely desired balance into disarray.
Motlop is averaging his lowest disposals, goals, contested possessions and marks in a season since 2012, and second-lowest inside 50s, scoring shots, marks inside 50 and goal assists in that time. Rockliff, in a severely injury-hampered role, has also been well behind his long term outputs, and this has drawn Port’s selectors one more player away from their ideal side.
Ryder’s injury cannot be understated in its impact on the structure of Port Adelaide. Last year he was the most valuable ruck in the league, and according to PAV one of the league’s best 10 players. With Port having no AFL-quality back-up rucks yet, the shuffling of key players like Dixon and Westhoff have created holes that are much harder to fill, and still have a downgrade of performance in the ruck as well.
There’s some logic to just playing one of these young rucks anyway, because instead of losing two players from ideal positions with any reshuffle, Port would just lose out at the original ruck position (this is before considering the importance of pure rucks in modern football). With Ryder back, this is somewhat academic, but surely something that Port should consider if and when he goes down next.
Watts, Motlop and Rockliff should all improve when roles elsewhere in the side are more comfortable and defined, and the coaching staff find the limitations of each player regarding their fit in structure of the team. Like the defence, we’d back the forward line to find their feet as the season progresses.
Even missing their three best midfielders, the Crows are keeping their head above water
If before the season you were to say that Bryce Gibbs, Cam Ellis-Yolman and Paul Seedsman were the Crows three best midfielders in terms of clearances and ball movement inside 50, you’d rightfully expect that the Crows would be in danger of missing the eight.
Not so fast.
Gibbs has long had a role, at least since the retirement of Chris Judd, holding together a weak midfield in an underperforming side. Now, he is doing a similar job at a better club – holding it together with duct tape and pride. Gibbs, according to PAV, has been the best first-year recruit at any club since Dangerfield joined the Cattery, and is on track for his best year ever at age 29. This is very unusual, and almost certainly will not hold through the rest of the season, but his performance through seven weeks has been heroic.
Ellis-Yolman might be an even bigger surprise. He was essentially chucked on the scrap heap by Adelaide and ignored by all 17 clubs at the end of last season before being re-rookied, and is now their second highest clearance winner. Several clubs could have used an Ellis-Yolman type, but none grabbed him. Greenwood, another free hit for the club in recruiting terms, is third in clearances for the Crows so far this year. In terms of midfield penetration, Seedsman has gone from supplying about 1% of Adelaide’s inside 50s a year ago to supplying 10% so far this year (best at the club).
Yet the Crows are winning the inside 50 battle against their opponents so far this year, which is remarkable considering Sloane and the Crouches have played 8 of 21 games between them. For their forward structure to get that scoring advantage as mentioned above, all the midfield movement need to do is break even, and they’ve been doing better than that. The clearances are down significantly, with the Crows winning less than opponents do, but not all teams live and die on winning that battle. When it comes to creating more scoring opportunities than their opponents, the Crows are more than keeping their heads above water.
Things could get scary for the league when they hit full health.
How the game might play out
One of the new things we’re testing this year is a prediction methodology called PERT (Player Estimated Results Tipping) which, distinct from Elo-based and other models, is tipping built from the ground up based on the selected sides, using Marginal Player Approximate Value (mPAV). It’s essentially a version of of the HPN Team Ratings built on the fly from the specific players in the team, rather than from the overall performance of the club.
On selected teams our prediction model PERT is predicting the Crows to win by just two points. The midfield ability of both sides is basically seen as even, and neither side should be able to generate many more forward thrusts than their opponent. At least on paper, that is.
What may swing things is the inclusion of a finally fit Rockliff and a non-suspended Powell-Pepper into Port’s midfield, but how Port integrates them is yet to be seen. As stated above, the Crows scoring ability against the Port defence should be a better battle to watch than the converse matchup, which are areas which both sides are about league average in. However, a match can be won from anywhere, and by almost anyone in modern football
Something we’re exploring with this season is assessing the impact a specific player has on the projected results of games. For instance, Marcus Bontempelli was a late out last week and we tweeted that him being replaced with Honeychurch reduced our prediction from a 16 point win over the Suns to a 10 point win over the Suns (it ended up being a 9 point result). As a little experiment, we’ve taken a look at the current selected sides for the Showdown and plugged in the quality missing players to see how things might have looked.
Later in the year, when both teams start to taper towards finals and hit September mode, we expect that the Crows have more talent to come back than the Power. Sloane (if the Lisfranc injury isn’t too bad), Brad Crouch and Brodie Smith (if he makes an unlikely return from his ACL injury) would likely improve the selected side by between two to three goals, depending on who they replace and their opponent.
By contrast, only Pittard, Marshall, Broadbent, and Hartlett (who is out for the year barring a miracle), are probable best-22 players missing for Port. None are as significant in nature, improving the team by around a goal combined. Like last year, Adelaide should be a better side than Port, but we still don’t know how good Port can be against a good side with all their structural issues sorted out. This game will answer some of these questions.
Psychologically, this game shapes as a test between two sides likely to face off for higher stakes this year, and it will provide some strategic insight to how they approach each other.
Get the popcorn ready.