This year HPN will do a brief club-by-club season preview of the upcoming AFL season. HPN will break down the teams both subjectively, and will predict how they will go on both our standard prediction method based on Pythagorean Expectation and Strength of Schedule, and a more detailed player based team analysis.
Following on from our work with Player Approximate Value (PAV) we have developed a per-game measure. The scores are based on performance above or below expectation. It is called Marginal PAV, or mPAV for short. Marginal PAV gives a positive or negative score based on the performance compared to what a perfectly “average” player would contribute in a game, which is roughly 0.66 PAVs per game based on a 22 game season by a perfectly “average” team. As better players play more games, this breaks down to the following:
| Average individual game mPAV (average marginal per-game PAV by definition) | 0.00 |
| Average player mPAV (average marginal per-game PAV for all players who played a game) | -0.06 |
| Replacement level (23rd best per-game marginal PAV on an average list) | -0.13 |
The average for all players in 2017 is -0.06 because that’s the average per-game output by every player who played a game. Better players play more games than the numerous fringe players, who only play a small number of games at a low level of value. More of the season’s total games are played by those who provide higher value and output so the player-average is lower in quality than the typical senior player-game which is by definition 0.00 mPAV.
Anyone with an mPAV of -0.06 or higher was more valuable per-game than half the league in 2017. Anyone whose mPAV is at least -0.13 is “replacement level”, good enough to be the “next in” for the average club, 23rd best on an average list. Most rookies don’t produce much value in this view, and quite a few viable prospects won’t hit replacement level, they’re there for longer term reasons.
To apply mPAV to a season long prediction, we have looked at the make-up of every list after the trade and free agency period, (including mature age draftees), and ranked the players with 2017 AFL experience from best to worst according to the measure.
From there, we have predicted the number of games each rank of player will play using data from the last three seasons, and applied a polynomial regression to smooth the data out. By multiplying the predicted number of games against each player’s PAV, we can create a predicted team strength for the upcoming season. Players who didn’t play in 2017, including new draftees, do not have a value in this system, and instead values are imputed for unidentified players based on the average of the bottom five players on a given list. Each side has 38 players predicted to play games, however this will not be the actual case for all clubs. Like all predictions, it will not be perfect, but it does provide a decent counterpoint to a team-based prediction.
Today, we will look at the first lot of clubs, from Adelaide to Fremantle. When we refer to a team being strong in an area of the field, we are often referring to the 2017 team strengths calculated below. These ratings also determine the scaling for total points in each team’s PAV for the year (so Adelaide have 110.2 PAV in offence).
Adelaide
Strength: That damn forward line.
The Crows were the most effective and efficient team in the competition up forward last year, and that looks to be even more of a strength in 2018. The Crows sat first in a number of key offensive areas, from points per inside 50 to marks inside 50 per entry to even raw scoring itself. That’s not to say that they are deficient elsewhere around the ground, with a above average midfield and a solid defence. You can pick your particular measure of offensive potency, but on almost all the Crows are at the top.
Weakness: Grand Finals. Too soon?
Compared to their formidable attack, the Crows midfield, and particularly their midfield depth in the second rotation, is a little bit weaker than the other parts of the ground on paper (well, maybe the defence, but we will get to that later).
Several of the more high profile losses that the Crows suffered in 2017 came at the hands of undersized, mobile forward lines. Melbourne caused a boilover in round eight, beating the ladder leading Crows by 41 points with a forward line anchored by targets such as Garlett, Bugg and Petracca. And in the Grand Final a small Richmond forward line decimated the tall timber of the Crows defence. The Crows may need to experiment with a smaller, quicker backline to counter such squads when they can foresee them.
The big question: Can they afford to keep replacing top end talent?
The (nearly) world beating Crows are coming back in 2018 with a couple of dents in the side of their bandwagon as mentioned above. Over the years the club has lost players like Tippett, Dangerfield and Vince, as well as those listed above, and has had to constantly replace their top end impact in their team. Eventually the cumulative impact of these losses will be felt; perhaps 2018 will see a slight slide as a result. The defence looks to have taken a hit with the departure of Jake Lever, and absence of Brodie Smith, this year. The Crows have several league average, or above-average, replacements available for the two players (Andy Otten, Curtly Hampton are just two), but how they will fit into the established Adelaide set-up is unknown. Lever played a very different role to both Otten and Hampton, and the Crows will likely have to make consequential adjustments to keep their heads above water.
The arrival of Bryce Gibbs from Carlton will provide the Crows with a extra top end rotation through the midfield, however a replacement for Charlie Cameron is not apparent from the displayed AFL performances from the current list. Gibbs has played at All-Australian level (or thereabouts) for the past few years at Carlton, and is able to go both inside and outside as a midfielder. However, with the aforementioned absence of Brodie Smith, maybe the Crows could most use him off a half back flank as a distributor?
Prediction: Pythag: 1st (82% chance of Finals). mPAV: 1st (85%).
Brisbane
Strength: Forward line potency
The Lions might not have racked up tons of big scores, but a lot of that had to do with an inability of their midfield to push the ball up forward rather than a lack of ability when the ball gets up there. The Lions are led by the finally no longer underrated Dayne Zorko, with Eric Hipwood one of the most promising key position prospects of recent years. The Lions have struggled to find a second key forward to slot in next to Hipwood, and the Schache experiment ended after just two years, but there’s already enough there to get excited about despite the presence of Josh Walker.
The Lions created the most scoring shots for every inside 50 last year, and were also top four for actual points per entry. Where they fall down is for marks inside 50 per entry – near the bottom of the competition, highlighting a potential need down the road. Or maybe they have worked out a forward plan that eschews the need for constant marking target. If that’s the case, they will pose a near unique problem for the rest of the competition to solve in the coming years.
Weakness: Everything else
Well, that might be a bit harsh. Stefan Martin is a solid mainstay in the ruck, and Archie Smith a promising backup, and the squad of outside midfielders is exciting (if they are ever all fit). But the rest is very much a work in progress. The backline last year had more holes than Swiss cheese, and losing Tom Rockliff from an already shallow inside midfield rotation looks concerning.
The Lions had the worst defensive team by almost any measure last year, and it’s hard to see what the plan from here is. Harris Andrews has some potential, and Luke Hodge should goes some way to filling a gap in the short term, but beyond that there are a lot more questions than answers. If the Lions can somehow stop the tide of inside 50s coming the other way, it should give the defence more time to think, but they may not get this luxury this year.
The big question: Are the kids alright?
Brisbane are not playing for 2018. They may not even be playing for 2019 or 2020. The goal for the Lions is a proper recalibration of their list, with the goal of a flag in 4+ years time. The club has been accumulating high draft picks and highly rated Academy players, along with young-ish players from other clubs, while working hard on culture improvement and retention. This is a good strategy, sort of a poor man’s GWS. Like early GWS, the results haven’t been there yet, but there are signs of hope scattered around; players like Witherden and Hipwood amongst others. Just don’t judge them too soon.
Prediction: Pythag: 18th (6% chance of Finals). mPAV: 15th (19%).
Carlton:
Strengths: Kicking the footy
Well, this isn’t much of a strength, more a style of play, but there is precious little to work with here for the Blues. The defence is serviceable at best, and Matty Kruezer has finally come of age as an elite ruckman (both tap and around the ground) in the competition, but the rest of the statistical indicators for the Blues indicate a struggling side.
So let’s focus on something a little more unusual instead; namely Carlton’s determination to move the ball by foot. Last season the Blues registered 4,965 kicks to just 2,948 handballs – or 1.684 kicks for every handball. The league average was 1.244, and no other side registered more than the 1.363 of Richmond and Port Adelaide.
Whilst those two sides were successful last year (the latter more than the former), it appears that Carlton are determined more than any other side to move the ball only by foot if possible. This may not be translating to any results as yet – but the Blues are also above average in another (somewhat) related metric, that of converting kicks to uncontested marks. Sometimes there is a significant advantage to radically altering the approach a side takes compared with the rest of the competition – whether Carlton continue down this path will be something to watch this year.
Weakness: Kicking goals
As alluded to above, the Blues have more than one weakness apparent, but let’s stick with the most obvious one – their forward line. The Blues finished stone motherless last for creating scoring shots per inside-50 last season, which is also where they finished in 2016. No team is more devoid of attacking power, with ruckman Kreuzer or journeyman Matthew Wright leading their charge up forward in recent years. At 27 it is unlikely that Levi Casboult will ever make the leap from “once-promising” to anything worthwhile, and Jed Lamb was their fifth-best forward in 2017 according to Marginal PAVs Per Game (mPAV). Let me pause for a second there…
JED FREAKING LAMB.
The slivers of hoper are the group of emerging 21 year olds, namely Harry McKay, Charlie Curnow and Jack Silvagni. If two of those three develop into proper options up forward (and perhaps one or two of Cunningham, Pickett or Petrevski-Seton), Carlton might be on the right track in a couple of years’ time. Maybe.
The big question: Where in the rebuilding project are they?
With the exit of Bryce Gibbs, it appears that the Blues are firmly in the middle of another rebuild, or deeper into the last one at least. However, for a rebuilding side, they appear to lean extremely heavily on their mature talent still. Beyond Cripps and Curnow, most of the young Blues talent still haven’t contributed significantly at AFL level, beyond short bursts anyway. Former number one pick Jacob Weitering seems to be pushed out of the best 22 backline by LIAM JONES, and might have to find a spot up forward instead. Marchbank, SPS and McKay, amongst others, have shown glimpses of becoming regular contributors to a good side, but they aren’t there yet. The question will be whether it is in the best interests of the Blues to continue to recycle their mature assets for draft picks or young players at the end of the season, or to keep some mature bodies around as mentors as the side progresses.
Prediction: Pythag: 15th (8% chance of Finals). mPAV: 16th (11%).
Collingwood
Strength: That first rotation midfield
When you have Scott Pendlebury, Adam Treloar and Steele Sidebottom leading the way, it’s hard to look another way. Along with Taylor Adams and Brodie Grundy, the Pies had three of the top 12 “Pure Midfielders” (defined separately from Mid Forwards or General Midfielders) in the competition.
| Player | Team | Off | Def | Mid | mPAV |
| Martin, Dustin | RI | 0.22 | -0.07 | 0.37 | 0.52 |
| Ablett Jr, Gary | GC | 0.12 | -0.02 | 0.38 | 0.48 |
| Sloane, Rory | AD | 0.08 | 0.04 | 0.33 | 0.45 |
| Pendlebury, Scott | CW | 0.07 | 0.01 | 0.33 | 0.4 |
| Cotchin, Trent | RI | 0.04 | 0.02 | 0.27 | 0.33 |
| Parker, Luke | SY | 0.04 | -0.03 | 0.3 | 0.32 |
| Selwood, Joel | GE | 0.04 | 0.01 | 0.26 | 0.31 |
| Kennedy, Josh P | SY | 0.03 | -0.05 | 0.33 | 0.31 |
| Wines, Ollie | PA | 0.03 | -0.07 | 0.34 | 0.3 |
| Adams, Taylor | CW | 0.01 | -0.03 | 0.31 | 0.29 |
| Treloar, Adam | CW | 0.03 | -0.05 | 0.31 | 0.29 |
| Kelly, Josh | GW | 0.09 | -0.04 | 0.23 | 0.28 |
If it wasn’t for this group of players, the Pies would be closer to fighting to avoid the wooden spoon rather than a spot in the eight.
Weakness: Tall Defenders
There’s fair reason to believe that the Pies’ attack will improve in 2018, as they were almost exactly league average for creating scoring shots per inside 50 entry, but well below average for converting these into goals. We’ve looked at this in the past, and this tends not to be an area where there are long term side-wide trends.
Instead, the state of the Pies defence seems to be an issue looking ahead. Whilst Jeremy Howe has been reinvented after his move back down back as a third man up/rebounding defender, the rocks behind him have been constantly shifting. Lynden Dunn, essentially an afterthought when recruited from Melbourne, has become a critical part of the defence, and this year Darcy Moore has been mooted as an option at CHB. Will it work? Probably not.
The big question: Will this finally be finals for Buckley?
The Pies were a part of finals last year if we are including breaking Melbourne hearts by keeping them out, but in a much more real sense they watched from the outer last September.
Prediction: Pythag: 9th (46% chance of Finals). mPAV: 10th (45%).
Essendon
Strengths: Kicking goals and taking names
The only side who created more scoring shots per inside 50 than Adelaide last year was Essendon, who were able to come back from their wasted 2016 with an incredibly solid performance up forward. Daniher transitioned from being one of the more promising tall forwards under 25 to perhaps the best, and Fantasia moved into fringe All Australian squad consideration. The Bombers were able to find ways to goal via both small and tall players, utilities and specialists alike. The key to beating the Bombers will be finding a way to slowing this attack down this year.
Weakness: The engine room
The Bombers struggled mightily both in the clearance and territory battle in 2017, holding the club back from potentially finding their way into the second week (and beyond) of the finals. Beyond Heppell, Merrett and Zaharakis, the quality of the Dons midfield dropped off significantly, and not all three of those players are “inside beasts”.
The big question: Where will the improvement in the middle come from?
The Bombers added three biggish names during the offseason in Stringer, Smith and Saad, with hopes that some or all of these three could help address their midfield issues. In at least Stringer’s case, that has not eventuated yet. Smith looks like a more likely inclusion in the middle, with his early (albeit preseason) work promising. And the recruitment of Saad could help open up time for bursts from Goddard through the stoppages. It’s unsure from our seats whether this will be enough to improve the deficiencies.
Prediction: Pythag: 8th (55% chance of Finals). mPAV: 7th (68%).
Fremantle
Strengths: Most Nat Fyfes in the competition.
Fremantle have truly hit a rough spot, just a few short years from playing in the Grand Final. Both the forward and back lines will be nearly unrecognisable from that squad, and the aspirations for the club are closer to escaping the bottom four than making the top. However, the Dockers still have a number of elite talents on their books, most notably former Brownlow Medalist Nat Fyfe. Fyfe is still young enough to see the prospect of another Dockers finals charge on the horizon, perhaps with fellow standouts Neale and Walters by his side. The Dockers actually broke even in the clearance battle last year, but after getting it out, struggled to get it to meaningful spots on the ground.
Weaknesses: How long do you have?
Let’s go with the tall forwards. In the 2016 trade period, the Dockers brought in Cam McCarthy in what looked like a steal for the club. The Dockers ended up nabbing two very promising players with the high picks from that trade (Sean Darcy and Griffin Logue), but even McCarthy cautioned against the optimistically high projections for his performance at the club.
McCarthy is a handy football player, but even at the Giants he wasn’t best used as a first option. Shane Kersten is similarly best used as a outlet option, and Freo have a plethora of smaller options but precious few talls. With the surprising emergence of Sean Darcy, there might be the opportunity to move one of he or Sandilands up forward to ease the pressure on options like McCarthy and Kersten. Otherwise, they will have to rely on their smaller brigade and midfield unit to manufacture a LOT of goals.
The big question: Can Lyon quickly build on the Dockers emerging core?
Of Fremantle’s top 10 players according to mPAV in 2017, eight are 28 or under in 2018, with four of those players under 25. Only the ageless Sandilands and David Mundy are over thirty in that mix, and will likely not be around for the next Fremantle golden period. The rest, however, should form a solid base for a charge to the finals in the next three to five years. The Dockers have added three top-10 draft picks in the last two drafts after not having done so since 2009 (the unlucky Anthony Morabito), and these players should develop whilst the likes of Fyfe, Walters, Neale, Wilson and Matera are still playing very good, if not great, football. The 2018 draft also presents another likely top 5 pick, and perhaps even the 2019 down the road. Like many of the clubs above, finals in 2018 might be less of a goal than playing for the future, but that’s no reason that they can’t build a solid gamestyle or footballing identity along the way.
Prediction: Pythag: 16th (7% chance of Finals). mPAV: 11th (27%).