Saturday, October 5, 2019

AFL Power Rankings: Finals 2019

As they did with their drought-breaking premiership in 2017, Richmond came home strong with close to their best line-up to be the premier team of 2019.

As many people (cough cough) predicted, Richmond emerged from the 2019 finals with their second premiership in three years. The Tigers won their three finals by a combined margin of 155 points. This made it 12 wins in a row to finish the year with an average margin of almost 40 points per game (and an average ‘adjusted’ margin of about that much as well – see chart below).


Injuries would seem to explain Richmond’s year … wouldn’t they?

Richmond was a far, far stronger team after its bye in Round 14, before which they had been about average. Their side had been hit hard by injuries in the first half of the year. Straight after the bye though they brought in seven players that ended up being in their premiership team: Trent Cotchin, Shane Edwards, Kane Lambert, David Astbury, Jayden Short, Shai Bolton, and Ivan Soldo. Later they also re-gained one of their main forwards Jack Riewoldt, and their main ruck Toby Nankervis.

It would seem then that the Tigers’ turnaround could be mainly attributed to getting more of their better players back. This was surely true to some extent: Cotchin, Lambert, and Riewoldt are clearly more productive players than Connor Menadue, Dan Butler, and Noah Balta. It is interesting to note though that Richmond saw a similar jump in form towards the latter part of their previous premiership season in 2017 (see chart below).


Injuries would seem to have been less of a factor for Richmond in 2017, as their top players generally missed no or few games. Like 2019 though, the Tigers may have progressively worked their way towards close to their best side – Richmond fans will well remember they kept the same line‑up all the way through the 2017 finals series. Jacob Townsend, Jack Graham, and Nathan Broad settled into the team late in the season, with Townsend kicking 16 goals in his five games. Also most of the players that had missed some time during the season, such as Dion Prestia, Nick Vlastuin, and Edwards, had returned by finals time.

Last year the Tigers came bang out of the gate, but had trailed off a bit by season’s end. This year Richmond seemed to just get stronger as the season went on. Even for the Grand Final this year the Tigers still seemed to be working their way towards their best line-up, with mid-season recruit Marlion Pickett famously making his debut in the final game of the year. There was one player though that had still not returned, and whose absence was thought by many to be terminal to the Tigers’ flag chances in 2019.

Covering Alex Rance

Back in Round 1 Richmond lost its best interceptor in Alex Rance to a season-ending injury. At the time, I suggested that the Tigers may be able to cover Rance’s contribution to some extent, but that he may still be missed. HPN had a similar message, noting that Richmond had become less reliant on Rance in recent years.

As it turned out, that is kind of how it worked out. Richmond averaged easily the most intercepts per match in 2018, with 82 per game. In 2019 this dropped to 75 per game. It was a bit better after the bye, increasing from 73 to 77 per game, which suggests that part of the difference came not from Rance but just from having better personnel on the ground.

I have written before how important intercept/turnover differentials are to scoring shot differentials with your opponent, as an intercept ends your opponent’s possession ‘chain’ and starts one of your own. The 2018 Richmond team took the strategy of deprioritising the clearances and more than making up for it through forcing turnovers to new levels. This year’s Richmond side was relatively bad at winning clearances as well, finishing third-last in clearance differential. Where then did the positive scoring shot differential with their opponents come from?

Richmond still had a pretty good intercept differential after the bye (see tables below), although not as good as last year. What Richmond did do better than last year was prevent their possession chains from ending up in stoppages.
On average after the bye they had seven less ‘chains’ per game stopped than their opponents, more than making up for the five less per game they started through clearances. Whether this was because their opponents had difficulty catching them, or they were still able to progress the ball when caught (or both) I don’t know, but it helped them make up for not being as heavy on intercepts in Rance’s absence.

Still, things didn’t look that good at mid-year. As a Tigers fan I’m glad they found a way to peak at the right time for the second time in three years. Now enough numbers … I’m off to go look again through the premiership memorabilia and wait for the Richmond beer to go on sale. See you next season!

Tuesday, August 27, 2019

AFL Power Rankings: Round 23 2019

There is no clear premiership favourite heading into the finals this year, but that will not stop the rankings making a pick anyway.


After the home and away season Geelong finish on top of the rankings, and the ladder, having spent the better part of the season on top of both. Below that though, things are quite close and muddled, as various teams have heated up and cooled off compared with their overall season form.

The Western Bulldogs are ranked second, but finished seventh on the ladder. Hawthorn and North Melbourne are ranked in the top eight, but did not make the finals. GWS and Essendon are ranked outside the top eight, but did make the finals. Last year’s Grand Final contestants Collingwood and West Coast are ranked outside the top six.

Good luck picking a premiership favourite heading into the finals this year. But we’re going to try and pick a flag winner anyway…

Predicting the finals series

Over the past four years I’ve used the rankings to predict how the finals series will play out. Basically this is done just by comparing who has the higher ranking points, adjusted for any estimated home ground advantage. This method has picked two of the four premiership winners (Hawthorn 2015 and Richmond 2017), and six of the eight Grand Finalists (missing out on the Bulldogs in 2016 and Collingwood in 2018).


This year, the rankings predict that the top two teams on the ladder – Geelong and Brisbane – will meet in the Grand Final, with Geelong prevailing (see table above). Indeed, the rankings are picking the higher-placed team on the ladder in every case, apart from the highly-ranked Bulldogs cutting a path through to a Preliminary Final.

Home ground advantage aside though, there is not a lot separating the four highest-ranked teams. Richmond is the current premiership favourite with the bookmakers, and the Tigers’ probability of winning could be better than indicated above. The rankings may still be underrating them a little due to their injuries earlier in the year, and an extra goal of ranking points would have them as favoured in any match-up outside of Brisbane at the Gabba (and the bookmakers favour them in that as well). That still would not make Richmond a strong premiership favourite however.

Outside of West Coast against Essendon, it would not be surprising to see the above scenario go completely out the window within the first week. Late season form ‘slump’ aside though, Geelong has still performed well enough throughout the year for the rankings to favour them as the premier team of 2019.

Sunday, August 18, 2019

AFL Power Rankings: Round 22 2019

The AFL tries to give the weaker teams the easier fixtures, and the stronger teams more difficult match-ups. However, teams rise and fall – who got lucky and unlucky in 2019 from their fixtures being a lot different than expected? (Hint: anyone who played Brisbane twice got unlucky.)


Whenever the AFL fixture for the new season is released there are several ratings of the difficulty of each team’s fixture, including on this blog. These ratings though are primarily based on how strong each team was in the previous season. In this post I re-rate the difficulty of each team’s fixture in 2019, based on how strong each team has been in this season.  

How the fixture was meant to work in 2019

In a season where each of the 18 AFL teams plays 22 matches, each team can only play five others twice. The AFL attempts to make the fixture more equitable by applying a ‘weighted rule’ to these ‘double match-ups’. Basically the 18 teams are divided into three groups of six based on how they finished after finals in 2018. Teams will tend to have more ‘double match-ups’ against teams in their own group, and less against teams at the other end of the spectrum.

This is who each team plays twice in 2019.


Ladder position sometimes hides the ‘true’ relative strength of a side. If we re-rate teams based on my rankings at the end of 2018 though the AFL did not seem to do too badly (see table below).


Fremantle were unlucky, in that they were the only bottom six side not to get to play Carlton or Gold Coast twice. My rankings considered North Melbourne and Sydney to be unlucky as well, as two of their ‘double match-up’ opponents – Geelong and Essendon – were rated more highly than their ladder positions.

(Note that, in the table below, combined ranking points have been reversed so higher points means an easier fixture. Also, points were adjusted so that their sum is zero.)

How the fixture has ended up working in 2019

Teams improve and decline, and so the strength of the fixture at the start of the season is different to how it actually turns out.

In 2019 the teams that have improved the most – based on their improvement in ranking points – include Carlton, the Western Bulldogs, and Brisbane (see table below). Facing these sides twice is now a tougher prospect than it first appeared. Geelong has jumped a lot of spots on the ladder, but the rankings consider that the Cats’ ladder position last season understated their ‘actual’ strength.


Conversely the biggest declines in performance, according to the rankings, have been from Melbourne and Essendon (despite the Bombers moving up the ladder). Sydney has dropped several spots on the ladder, but the rankings do not consider its fall to have been as great.

If we re-rank the strength of each team’s ‘double match-ups’ we see some significant changes for some teams (see table below).


Almost every team that played Brisbane twice is now considered to have had a fixture difficulty of a ‘top six’ side. The exception is Gold Coast, and its fixture difficulty has still moved from that of a ‘bottom six’ to a ‘middle six’ side.

Top sides Richmond, Collingwood and West Coast are now considered to have had much weaker fixtures than originally intended, as they played Melbourne twice. Richmond and West Coast’s fixture difficulties are considered to be closer to ‘bottom six’ or ‘middle six’ sides.

Sydney has gone from unlucky to not as unlucky, helped by playing Melbourne and Essendon twice.

GWS did not play Melbourne twice, but every team it played twice except Gold Coast is rated at least a bit weaker than last season.

Ultimately the fixture doesn’t turn you from a ‘good’ side to a ‘bad’ side, or vice versa. Further, this season the variation in fixture difficulty across teams is less than usual – there are no very strong teams, and with Carlton’s improvement there are no very ‘weak’ teams except for Gold Coast. Still, as Richmond and West Coast are jostling for top four positions, they may be a little bit thankful that their return bouts against Melbourne were a lot easier than they looked at the start of the season.

Sunday, July 28, 2019

AFL Power Rankings: Round 19 2019

The Brisbane Lions have been near the top of the ladder for most of the season, but it is their recent form that has made them a premiership contender.


From ‘cellar dweller’ to top four

If you are reading this, you probably already know that the Brisbane Lions have jumped up the ladder in 2019. In 2018 the Lions had only five wins and finished fifteenth. In 2019, as of the end of Round 19, they have twelve wins and are in third. No team has moved up nearly as much as the Lions have this season.


If you look a bit closer at the Lions’ results in 2018, they were actually better than their five wins would suggest. Their percentage was 89.1, a fair bit better than five of the teams in 2018. Their average net margin, adjusted for estimated home ground advantage and opponent strength, was ‑11 (see chart above); certainly below average, but not horrible. The Lions’ poor win-loss record in 2018 in part came from them winning only one out of their six games decided by less than seven points. A team with their performances would normally be expected to win between eight and nine games.

The Brisbane Lions were therefore expected by many analysts to improve. Few expected them to improve by quite this much though. Outside of their ‘purple patch’ in Rounds 15 to 17 last year, they still were one of the weaker teams. But just a few rounds out from the end of the home-and-away season this year, the Lions are suddenly a premiership contender.

Bolstered by the fixture, boosted by a burst of form

A few weeks ago, this blog – like many others – did not yet consider Brisbane as one of the main premiership contenders, despite the Lions sitting in the top four at the time. These rankings had them as the ninth-best side, even though they had been outside the top eight after only one round all season.

Over the first half of the season, however, Brisbane had arguably only been slightly above average. From Rounds 1 to 12, their average adjusted net margin was only +3 (see chart below). The Lions’ record up to then has been somewhat bolstered by a ‘friendly’ fixture. They have yet to play Geelong or Richmond, and they didn’t play GWS until Round 16. They played West Coast and Collingwood at home – impressively beating the former, and getting well-beaten by the latter. They were well-beaten by Essendon. They lost to Carlton in Round 12. Brisbane had certainly improved, but they had been helped a bit by beating the ‘lesser’ teams.


In their past six matches though, the Lions have hit the stratosphere. Their average adjusted net margin in that time has shot up to +33 (see chart above). They easily beat St. Kilda and Melbourne in Rounds 14 and 15. They beat GWS away in Round 16. They smashed Port Adelaide away in Round 17. They had a close win against North Melbourne in Round 18, but registered 12 more scoring shots. Then on the weekend they comfortably beat Hawthorn in Tasmania.

Brisbane is now fifth on the rankings, with two goals worth of ranking points, and they are less than a goal from top spot. In a year without a strong team, the Lions are now a genuine ‘premiership contender’.

Bolstered by the midfield recruits

After the 2018 season, Brisbane lost former captain Dayne Beams to Collingwood. They did however manage to get star midfielder Lachie Neale from Fremantle. They also recruited Jarryd Lyons, who was amazingly delisted by the Gold Coast Suns despite being arguably their most productive player.

The Lions have significantly improved this year in terms of contested possessions, clearances, and inside 50s (see table below). Their midfield recruits have led to some of this improvement. Despite being far less heralded than Beams, Lyons has basically replaced his production. Meanwhile, Lachie Neale – who is fourth in the league in contested possessions per game, and second in clearances – is a massive upgrade on the Lions’ next best midfielder, Tom Cutler. Neale and Cutler play different roles, but Neale’s arrival has led to some reshuffling of roles for other players like Mitch Robinson, who offers more than Cutler did. Some of Cutler’s production on the outside has also been made up by the major improvement of Hugh McCluggage, who has been one of the best wingmen in the league this year.

Again, one might have expected some improvement in Brisbane’s midfield this year, but probably not to this extent. We will see if it can hold up through the months of August and September.

Sunday, July 21, 2019

AFL Power Rankings: Round 18 2019

Carlton has become the second team this season to significantly improve after a change of coach.


A few weeks ago, this blog showed how much North Melbourne had improved after its coach Brad Scott left, and was replaced by Rhyce Shaw. Another team replaced its coach this year when, after its Round 11 loss, Carlton sacked its fourth-year coach Brendon Bolton. Carlton had been awful in 2018 and in the first 11 rounds of 2019, winning just three matches in that span. Like the Kangaroos, Carlton named a ‘caretaker’ coach – David Teague – for the rest of the season.

Carlton’s transformation in its six matches since has been about as dramatic as that of the Shaw-led Roos. The Blues have won four out of those six matches, and lost the other two matches by less than a goal. While Brisbane is the only team they have played in that stretch that is in the top eight, being able to beat the lower sides is still a major improvement compared to where they were at. Even after adjusting for the strength of their opponents, Carlton has performed more like a side that is about average (see chart below).


Teams scored on average over 100 points a game against Carlton last season. In the Blues’ two most recent losses, the other team has also scored over 100 points. In their four wins however their opponents have scored no more than 75 points. Also, except for against Gold Coast, Carlton has had more inside 50s than their opponents in all of those wins.

The improvement in form could be attributed to the development of a young side. Actually though, it is probably more that the ‘older’ players have performed better. Ed Curnow (29 years old) and Marc Murphy (31 years old) have upped their contested possession count in recent weeks. Matthew Kreuzer (30), Kade Simpson (35), Levi Casboult (29), Liam Jones (28), and Nic Newman (26) have also been important to the turnaround. By using this mature core well, Teague now has the makings of a half-decent side.

An article on the ABC News site last week showed that being an AFL coach is actually a relatively secure professional coaching job. Does the improvement in North Melbourne and Carlton this season suggest that clubs should look at changing coaches more often? Should St. Kilda – also winners on the weekend – expect some improvement now that it has parted ways with its coach Alan Richardson? The cost of an expensive pay out aside, it is not clear that, on average, changing a coach or manager leads to any more improvement than if the incumbent had stayed on.

Carlton though was in the position where things could not have been much worse. David Teague has at least given the Blues some definite improvement in the short-term; whether this translates to improvement in the long-term remains to be seen.

Sunday, July 7, 2019

AFL Power Rankings: Round 16 2019

Of the main premiership contenders West Coast is in good form, Collingwood has fallen out of form, Geelong and GWS may be stuttering slightly, and we are still getting a handle on a ‘full-strength’ Richmond.



A premiership race in five, that became three, is five again

About nine or ten rounds into the 2019 AFL season there seemed to be five main premiership contenders: reigning premiers West Coast, 2017 premiers Richmond, and ladder leaders Geelong, Collingwood, and Greater Western Sydney.

Then in Round 12 it seemed more like a ‘race in three’. An injury-depleted Tigers got destroyed by Geelong, West Coast were well beaten (by Sydney), and Collingwood had a strong win in the Queen’s Birthday match.

Move forward to Round 16, and West Coast and Richmond are considered among the leading flag contenders again. Both won by over 90 points on the weekend. With TAB, the Eagles and Tigers – along with Geelong, GWS, and Collingwood – are shorter than $10 to win the premiership, and no other team is shorter than $20.

West Coast is rising, Collingwood is falling, and we are not clear about the Tigers

Of the five main premiership contenders, West Coast seems to currently be in the best form. Its average net margin over its past five matches, adjusted for estimated home ground advantage and opponent strength, is +26 (see chart below), even better than Geelong’s +18. The Eagles did get somewhat lucky by the Dockers scoring only two goals from 21 scoring shots on the weekend. On the other hand, the Eagles scored only 14 goals from 36 scoring shots against Essendon two weeks before.


Collingwood has been a below average team over its past five matches. The Magpies have lost to North Melbourne, Hawthorn, and Fremantle, and had a narrow win against the Bulldogs. It may be just a temporary slump, or it may be that their ‘peak’ ends up having been the 2018 finals, and the first few rounds of 2019.

Richmond has been a below average team over its past five matches as well. The Tigers had three fairly horrid losses in that stretch against North Melbourne, Geelong, and Adelaide. After their bye though they had several of their best players return, and have performed more like the strong side we saw in late-2017 and 2018. Richmond’s average net margin over its past three matches – even adjusted for the weakness of its past two opponents – is +24 (see chart below).


Meanwhile Geelong has lost two of its past three matches, and has a three-round average adjusted net margin of just +1. GWS has also lost two of its past three, and its three-round average is +3.

Before its recent drop in form though the Cats did clobber Richmond, and are still easily the highest-ranked team. They may be stuttering a little, but if I had to name the team I thought was most likely to win the 2019 premiership, Geelong would be it.

Monday, July 1, 2019

AFL Power Rankings: Round 15 2019

North Melbourne has improved its performances considerably in the four matches since Rhyce Shaw became coach, but it is not quite the season’s most ‘out of the box’ month-long stretch.


From Brad Scott to Rhyce Shaw and upwards

Around the time of North Melbourne’s Round 10 match against the Western Bulldogs, coach Brad Scott – the team’s coach for almost a decade – decided to leave the club mid-season. Despite their win against the Dogs, the Kangaroos at the time had only three wins from ten matches. Following Scott’s departure assistant coach Rhyce Shaw was named as the ‘caretaker’ for the rest of the season.

Since Shaw’s appointment, North Melbourne has gone from a below average team to one of the league’s better sides. It has beaten two of last year’s top four teams – Richmond and Collingwood – by around 40 points each. It also comfortably beat the Gold Coast Suns, with its only loss coming to GWS. The Kangaroos have surged on the rankings, gaining two goals worth of ranking points, and jumping from fourteenth to seventh.

This turnaround must be due to Rhyce Shaw, right? Give him the permanent coaching gig! Well, it could be the case that Shaw’s coaching is largely responsible for turning the Kangaroos’ performances around. But four-week stretches such as these, where a team performs significantly better than it has done for the rest of the season, are not unheard of.

The Kangaroos’ hot form is unusual, but not overly so

North Melbourne’s recent jump in form, compared to its overall performances for the season, is arguably not even the ‘hottest’ four-round stretch this season. Based on my adjusted net margins, that could be St. Kilda during Rounds 2 to 5 (see table below).



The Saints were not as good as North have been, but compared to their overall season form (not great) it was a pretty good stretch. They beat Melbourne easily, beat Essendon and Hawthorn, and came within a goal of beating Fremantle playing away. Since that stretch though, sixth-year coach Alan Richardson has come under ever-increasing media scrutiny about his job security.

Essendon also had a similarly improved stretch during Rounds 3 to 6, when they thrashed North Melbourne and Brisbane, beat Melbourne, and almost beat Collingwood. Their coach, John Worsfold, is in his fourth year at the club. And the other team I rate as having a big jump in form is Sydney during their recent climb up the ladder, with a coach – rumoured Kangaroos coaching target John Longmire – that has been in charge for almost a decade.

In the end though, it’s not the improvement for North Melbourne that matters most, but the level that it is playing at. The Kangaroos’ average adjusted net margin over their past four matches is over four goals. If a Rhyce Shaw-led side can consistently play at that level, then North Melbourne should be more than happy with where they are at.

Sunday, June 16, 2019

AFL Power Rankings: Round 13 2019



If my rankings' predictions were being used on Squiggle, I'd be at the top of the leaderboard. Just saying guys. :)

(Of course, if they were near the bottom, I'd be dead quiet about that...)

Monday, June 10, 2019

AFL Power Rankings: Round 12 2019

Nothing will ever diminish the West Coast Eagles’ 2018 premiership triumph, but they are far from the most fearsome ‘reigning premier’ we have had.


The reigning premier is clearly not the best team so far

The title of ‘reigning premier’ deservedly carries a lot of weight. Like a boxing heavyweight champion, the team that won the premiership last season is often season as the ‘team to beat’, up until the point that they are eliminated from the current season’s finals race.

In 2019 the title of reigning premier belongs to the West Coast Eagles. Halfway through the season the Eagles are hardly looking like the fearsome ‘team to beat’. They sit fourth, with eight wins out of 12 matches, three wins behind pace-setters Geelong.

More worryingly, they have a percentage of only 103.0. The Eagles have been well-beaten three times this season by teams that are merely around average: by Brisbane in the opening round by 44 points, by Port Adelaide at home by 58 points, and this weekend by Sydney by 45 points. (They were also beaten by Geelong by 42 points.)

West Coast’s average net margin this season, adjusted for estimated home ground advantage and opponent strength, is only +3 points (see chart below). Clearly, the Eagles to date in 2019 are not the most dominant reigning champion we’ve had.


The Eagles were not a dominant premiership team

A premiership is a premiership, and always deserved. West Coast was fantastic in last year’s finals series, with an average adjusted margin of +37 points (see chart above). The Eagles beat Collingwood twice – including once at the MCG – and destroyed Melbourne in the preliminary final. When fully fit, they may have been the best team for the year.

As far as premiership teams go, the 2018 Eagles were hardly the most dominant (see table below). They were probably similar to the 2017 Tigers, and slightly better than the 2016 Bulldogs. According to the rankings however, every other premiership team this decade was considerably better.

Further, last year’s West Coast team may not have even been the strongest Eagles team this decade (see table below). I’d say that was the 2015 Grand Final team, that had a percentage (after finals) of 142.8, but who ran into the triple-premiership winning Hawthorn side. Arguably the 2011 and 2012 sides were about as good as last year’s premiership team as well – their ranking points and percentages were similar.


The 2018 Eagles team benefitted from not having a ‘great’ opponent, or even too many ‘very good’ sides, in its way. (Richmond was possibly great early in the season, but had fallen off somewhat by the finals series.)

The low-rated midfield

Champion Data somewhat controversially rated West Coast’s playing list as only the eleventh strongest heading into the season, and its midfield way down in fifteenth. That may seem laughable to some who saw Luke Shuey and Dom Sheed tearing it up on Grand Final day.

As highlighted by HPN a few weeks back though, the Eagles are pretty low down in terms of their inside 50 differential with their opponents this season (currently fifteenth). They also rank low in differentials for disposals (sixteenth), and centre clearances (last). They’re not exactly lighting it up on the main midfield indicators.

Champion Data rated Luke Shuey, Elliot Yeo, and Andrew Gaff as above average midfielders. However, it rated Dom Sheed, Mark Hutchings and Chris Masten as below average (that rating may have changed for Sheed this season). In other words, Champion Data rated the Eagles as having little midfield depth, and no player among the very best midfielders – not even Gaff, who gets a lot of possessions, but who also has a low average number of metres gained per disposal for a wingman. (Shuey, last year’s Norm Smith Medallist, was of course ‘elite’ on the day that mattered most.)

West Coast is still a good side. After each of their three premierships the Eagles have been a finals side the next season, but not a huge premiership threat. This year – on their form – doesn’t look like changing that pattern.