Hawks year and review and line comparion to top 10 teams
I'm on holidays and had some time to kill, so I thought I'd pull together a bit of a statistical year in review of the Hawks.
I pulled the key publicly available stats, allocated players to positions/roles, and then rated each player based on the stats that I thought were most important for that role. For example, intercept marking carries more weight for a key defender, while goal-kicking accuracy matters more for a key forward.
I then compared our players and lines against players from the other top 10 teams.
A few caveats: this is very much my own model rather than an official AFL rating system. Players don't always fit neatly into one position, and forcing them into roles inevitably introduces some subjectivity. I also forced three players into each line, except for ruck (one player) and outside mids/wings (two players), giving me an 18-player "best" group.
The percentages below are essentially a measure of where that group sits relative to the other teams/players in that role, rather than a conventional score out of 100. So this is a percentile score. So for example for the ruck line for Melbourne is 100% as its only one person and Gawn is marked as the best ruck.
The rankings are a bit different and rank the line out of ten compared to other top 10 teams.
Anyway, I thought the results were pretty interesting.
Small forwards: 1st — 95th percentile
This probably won't surprise anyone: Hawthorn has the best small-forward group in my model.
Nick Watson was the highest-rated small forward, while Dylan Moore and Connor Macdonald also rated at an elite level. Even Jack Ginnivan, who was our fourth-highest-rated small forward, still came out in the top 20% of small forwards across the league.
The best three averaged around the 95th percentile in my model.
Brisbane was second at 87%, with Melbourne third at 86%.
This is probably the clearest strength on our list. We have multiple players who can hurt teams inside 50, and importantly, they aren't all doing exactly the same thing.
Key forwards: 3rd — 77th percentile
This was another strong area for Hawthorn.
Jack Gunston ranked second among the key forwards in my model and averaged 3.5 goals per game up from 3.1 last year.
Chol and Lewis both came out slightly above average.
The interesting caveat is that both Chol and Lewis tended to produce their better games against weaker opposition, although Lewis' preliminary-final performance was a pretty obvious exception.
Adelaide had the best key-forward group at 85%, with Fremantle second at 81% and Hawthorn third at 77%.
So while our forward line is clearly one of the strengths of the team, the numbers suggest that the small forwards are genuinely elite, while the key-forward group is more good than dominant.
Inside mids: 9th — 49th percentile
This is probably the most interesting result.
All year there was a lot of discussion about Hawthorn's midfield being our biggest weakness, and the model does support the idea that it's an area we need to improve — but it doesn't suggest we were an absolute disaster either.
Jai Newcombe rated in the top quarter of inside midfielders, which is particularly impressive given how strong that group is (14 of the top 15 Brownlow vote winners are inside mids, with Gawn the only exception). He was third in the league for clearances per game at 7.4.
The issue is that his possession numbers (25 per game) and disposal efficiency (73%) were below the very best inside mids.
The good news is that all of those numbers were significantly better than last year.
Josh Ward was slightly below average, although his rating and basically all of his underlying numbers improved significantly from last season and was elite is some key games and moments.
Will Day rated below this group as well. His rating was hurt by some poor performances in important games, including the preliminary final and games against Collingwood and Brisbane. I'd expect him to rate considerably higher next year if he can put together a full pre-season and avoid major injuries.
Sydney had the strongest inside-midfield group in my model at 81%, followed by Carlton at 74% and the Bulldogs at 67%.
So I wouldn't describe our midfield as poor. I'd describe it as the area with the biggest gap between where we are and where the best teams are.
Outside mids / wings: 9th — 14th percentile
This was easily the ugliest result for Hawthorn.
One reason this may not have been as obvious watching us is that we often use our fleet of small forwards to push up the ground and effectively fill some of the outside roles. That worked pretty well for much of the season.
But when I looked specifically at the players whose primary role was outside midfield, the numbers weren't great.
Massimo D'Ambrosio was solid, but there wasn't much else to get excited about statistically.
Harry Morrison ended up in this group and had some good games, but also went missing at important times. We also rotated players such as Josh Weddle, Karl Amon and even James Sicily through this role at various points, but I didn't think any of those combinations worked particularly well.
Collingwood had the best outside-mid group at 74%, with Sydney second at 72%.
This is probably the biggest statistical hole on the list.
Rebounding defenders: 2nd — 87th percentile
This was another excellent group.
Jarman Impey, Blake Hardwick and Karl Amon all rated highly.
Impey and Hardwick in particular made big improvements across a number of statistical measures. Impey gained roughly another 100 metres per game and four disposals, while Hardwick improved his disposal efficiency to around 84% and added about three possessions per game.
Amon was down on his numbers and overall impact compared with last year, but still had a solid season.
One caveat here is James Sicily. I classified him as a key defender, but you could very easily put him into this group, which would make an already strong line look even better.
Fremantle ranked first at 93%, with Adelaide third at 71%.
Key defenders: 1st — 70th percentile
Another strong result, although this is one where the headline number needs a little explanation.
James Sicily, Josh Battle and Tom Barrass made up this group in my model.
Battle was actually slightly up across his statistical categories from last year. Given that he was All-Australian last year, I thought he was a little unlucky to miss out this time around.
Barrass didn't have as good a statistical year as last season and was down across a number of categories. I also thought there were a few occasions where opposition forwards were able to expose him for leg speed, particularly in important games.
The Hawks still ranked first among the top 10 teams for this group, however, with Sydney second at 62% and Brisbane third.
So even with Barrass having a less impressive season statistically, the overall group remains one of the strengths of the side.
Ruck: 8th — 69th percentile
Lloyd Meek had a decent year, although his numbers were down across the board.
The changes to the ruck rules may have played a part here, although I wouldn't read too much into that without looking at the specific impact of the rule changes in more detail.
Melbourne, unsurprisingly, had the strongest ruck group with Gawn, while Fremantle's Jackson ranked second.
Meek isn't a major concern for me, but he's another position where the very best teams have a significant weapon.
So what does it all tell us?
Overall, I think the numbers paint a pretty good picture of where Hawthorn is at.
Our forward line is genuinely strong, particularly the small forwards. Our defence is also very good. The midfield is improving, but still has a gap to the best teams — and the outside-midfield numbers are ugly compared to peers.
That actually lines up pretty well with what we saw on the field.
The really encouraging thing is that some of our younger midfielders improved significantly, while players like Impey and Hardwick also took another step.
The less encouraging part is that our success still relies heavily on a few areas of the ground being excellent. If we want to go from being a very good team to a genuine premiership contender, the midfield — particularly the outside running — looks like the obvious area where improvement could have the biggest impact.
And then there's succession planning.
Gunston has been phenomenal, but obviously isn't going to be around forever. The same eventually applies to some of the older defensive players. We therefore need to keep developing the next generation while the current group is still good enough to contend.
The good news is that this seems to be broadly what the club is already trying to do.
Anyway, that's my very amateur attempt at turning a holiday afternoon into a statistical analysis of the Hawks.
I'm sure there are plenty of holes in the methodology, so fire away.