Statistical Analysis of Patrick Dangerfield

[Yoda_]

You make a lot of mentions to his lower position rankings in various stats but sometimes that kind of data can be misleading when you don't know the actual value difference between those rankings.

e.g. (just pulling numbers here) Dangerfield might be ranked 80th in some stat while other midfielders are ranked around 20th but what isn't shown is that the actual difference between 20th and 80th is only an average value of 0.8, so in reality there isn't really much of a difference there.

Not saying you're trying to mislead people with how you're presenting stats but anyone who looks at it objectively might question why you wouldn't actually include the totals and averages there. Unfortunately for you, being a Port Adelaide supporter in this case, people are going to assume you're pushing an agenda.

also comparing player x to player y (who is the best in that stat) is also similarly misleading.

 

[Rise_up_Stiffy]

Yeah Danger sucks pls don't offer him any money we will endure his poor football ty

 

[clogged]

Well, I'm not saying he's a master tackler, but at the same age Dangerfield is now, he was in the top 10 tacklers.

http://afltables.com/afl/stats/2006s.html#sh7

As the OP has pointed out, Dangerfield has never even been in the top 100 tacklers. That is poor for a guy who spends the majority of his time on the ball.

Damon_3388 your rebuttal? Inconsistent to unprolific? I would have thought top 10 in the league was pretty prolific.

 

[kiwicrow]

It would be an understatement to say that the other Dangerfield thread isn't exactly full of objective analyses of his strengths and weaknesses as a player. So, I am starting a thread to rectify that. What I have done is compared Dangerfield with many of the top midfielders in the league over the past four years across a range of key metrics. The other players are Ablett, Swan, Pendlebury, Griffin, Johnson, Selwood, Mitchell, Watson, Beams, Cotchin, Kennedy and Sloane.

This is not meant to be a complete list of the best mids over that timeframe and so don't be offended by individual inclusions or exclusions (Judd, Murphy, Boyd, Jones, etc). My aim was simply to see if Dangerfield could be thought to be in the top-5 among a particular group of midfielders that many would regard as the competition's benchmarks over the period. I threw Sloane in because I wanted to compare Dangerfield to Adelaide's other current top midfielder.

Two caveats before I start. First, the indicators I focus on - total disposals, effective disposals, contested disposals, the variance of total disposals, goals, goal assists, tackles, and inside-50s - do not fully summarise what a player adds to a team. I only had so much time to put this together and am happy to see others add other indicators that they think are relevant to the discussion (1%ers, metres gained, hard ball gets, etc). Second, yes, I am a Port supporter. However, as you can see from the rarity of my posting, I'm not involved in the daily BF flame wars. So, judge me on my analysis, not the team that I support.

Let me begin with Dangerfield's relative strengths. First, he is arguably the best goal kicking midfielder in the game. In three out of the last four years he has been in the top-100 players in the competition for average goals per game. Ablett, Swan and Beams have also been in the top-100 three times but Dangerfield's goals per game is slightly higher. Steve Johnson has a higher average over the past four years but that is driven by his 2011 numbers. Since he became more of a full-time mid his numbers have dropped below Dangerfield's averages.

Dangerfield's value forward of centre can also be seen by comparing his inside-50 stats. In 2012 he was ranked second in the competition for average inside-50s per game and in 2013 he was eighth in the competition. He is down this year but it is still early in the season and so the sample is too small to make strong inferences. Other players that excel in this area are Griffin, Ablett, Swan, Selwood and Johnson, though Dangerfield's averages over 2012 and 2013 are second only to Griffin over that time frame.

Dangerfield's other key strength is in contested posessions. He was ranked second in the comp for contested posessions per game in 2012, sixth in 2013 and fourth so far this year. Indeed, over that three year period his average ratio of contested posessions to total posessions has been 60%. The only top mid with a better ratio is Kennedy from the Swans and even then only just. For comparison, Ablett and Pendlebury both have ratios below 50%. As an interesting aside, in 2011 Dangerfield was ranked 32nd in the comp for contested posessions despite not even being in the top 100 in the competition for total posessions.

Now for the weaknesses. First, Dangerfield simply gets less of the ball than the other top midfielders in the game. His best result for average posessions per game over the past four years is his ranking of 14th in 2012. He then dropped to 47th in 2013 and is 61st so far this year. He averages 8 fewer posessions per game than Ablett over the past three years, six less than swan, 5 less than Pendlebury, 4 less than Mitchell, 5 less than Beams, 4 less than Watson, and the list goes on.

Unfortunately, he does not appear to make up for getting less of the ball by using the ball more effectively. His best ranking for total effective disposals is 48th in 2012 and he was ranked only 99th last year. His average ranking decline between total disposals and effective disposals is larger than for any of other top midfielders in the competition besides Cotchin. This is where his reputation for burning the ball comes from. Other players whose ranking slides on the basis of effective disposals are Johnson, Swan and Kennedy. Interestingly, Sloane's rankings improve on this metric - he has been ranked in the top 100 for effective posessions in each of the past three seasons.

Now an important reason why Dangerfield's effective possessions are down is because a larger proportion of his posessions are contested. But the average ranking decline is seems larger than can be explained by that alone. For example, Kennedy's average ranking decline is 15 places compared to more than 40 for Dangerfield.

Another area where Dangefield appears to fall short is in goal assists. He has never ranked in the top 100 in the competition on this metric. Every one of the other mids in my comparison list have done so on multiple occasions, with Ablett, Pendlebury and Johnson doing so in each of the four years. Sloane has ranked in the top 100 twice. So, Dangerfield scores more goals than other comparable mids but this is partially offset by the fact that he contributes less to others. I don't watch enough of his games to judge whether that fact is attributable to selfishness or to other factors.

Finally, Dangerfield does not appear to be a committed tackler. The only year of the four that he has been in the top 100 in the competition for average tackles per game is 2014 and even then he only ranked 90th. Again, almost all the other top mids are higher up in the tackle rankings. Ablett, Pendlebury, Selwood, Kennedy and Sloane have been ranked in the top-100 every year. The only other top mid not to appear in the top 100 in any year is Swan. I was surprised by this finding as usually there is a high correlation between contested ball winning and tackling.

In summary, what you get from Dangerfield is one of the best contested ball winners in the competition, who drives the ball into the forward line and scores more goals than other top midfielders. If you were judging the best mids in the comp on that basis alone he would probably be ranked top five. However, those positive attributes are somewhat offset by the fact that he doesn't get as much of the ball as the other best mids in the game and disposes of it much less effectively. While his high contested posessions point to a strong team player, his low tackle count and goal assists may be suggestive of a selfish streak. He also has a high variance in his posessions from game to game compared with many other mids.

On balance then I think it is difficult to argue that he is currently one of the top-5 mids in the competition, though he is clearly a great player that would add a lot of value to any team in the competition, including Port! To get there I think he needs to raise his effective disposals and get more involved in tackling and giving off goals to others. Beams comes out better on those categories, though not in some others.

Of course, he is only 24 and so probably has not reached his peak yet (despite the fact that 2012 was his best statistical season so far). Nearly all of the other mids I identified are older and will start to drop down the rankings over the coming years. The question is whether Dangerfield will rise up or other, even younger mids, will overtake him.

Or to rephrase the above>
"Even though my team is doing really well both on and off the field right now, I'm so miserably absorbed by the AFC that i can't help but put up a bit of BS attacking a player that seems like a pretty decent bloke and a hard at it footballer"

 

[Ultimatt]

Good effort OP. I definitely appreciate the work you've put in to it. The tricky thing about doing a stats analysis in AFL is giving value to those stats. Compare it to the big american sports where it's pretty simple: A running back is judged on metres gained, a point guard on field goal %, a baseballer on batting average. The easy and most common stat for AFL midfielders is disposals but that doesn't tell us much: Was it effective? How far did it go? Handball or kick? Was it an attacking kick i50 or crossing out wide? While the numbers can give us a rough idea of how good a player is, they only go so far. On top of that, there's 18 players out there with more varied roles/positions than what you'd see in American sports. It's hard to compare an inside mid to an outside mid and then you need to consider how much time they spend around the ball compared to time forward/behind the ball.

Having said all that, your stats do give some good insight into the strengths and weaknesses of Dangerfield and compare those to other players strengths/weaknesses, but I don't think you can take that further and compare who is the better player just off the numbers.

 

[Mr_Moogle]

Dangerfield is a beast. Any club in the land would kill to have him on their list.

 

[Dandy_GO]

It's a complete and utter waste of time to conclude that a guy who clearly spends far more time up forward than most pure midfielders isn't a "top 5 mid" on the basis that his stats don't match up to those who spend more time around the ball.

You wrote a lot of words to come to a blatantly flawed conclusion.

 

[Damon_3388]

Damon_3388 your rebuttal? Inconsistent to unprolific? I would have thought top 10 in the league was pretty prolific.

Now lets look at his year-by-year tackling numbers, and see how consistent they are.

 

[clogged]

Now lets look at his year-by-year tackling numbers, and see how consistent they are.

Consistency year to year? What are you on about? Players assume different roles depending on the year, and injuries. His coach may has asked him to play more of an outside role one year versus the next. What matters is consistency game to game.

What we do know is that Dangerfield has never been a big tackler. Ever. The comparison you made is ridiculous.

 

[Liverpower]

It's a complete and utter waste of time to conclude that a guy who clearly spends far more time up forward than most pure midfielders isn't a "top 5 mid" on the basis that his stats don't match up to those who spend more time around the ball.

You wrote a lot of words to come to a blatantly flawed conclusion.

Actually, a number of the indicators I put forward ought to advantage someone that spends more time up forward. For example, I included inside 50s but not rebound 50s as one of my metrics. I also included goals kicked and goal assists. Both of those ought also to favour someone like Dangerfield. I do think that it is somewhat concerning that his goal assists are so low - don't you? I mean, should he not have more opportunities given that he spends so much more time up forward than the other players he is being compared to?

And if you read the piece less defensively, I made it clear that the analsysis is consistent with him being an extremely good and valuable player. It is just that people take that too far and want to claim that he is one of the top 5 players in the competition. That is an exagerration in my view.

 

[Liverpower]

Good effort OP. I definitely appreciate the work you've put in to it. The tricky thing about doing a stats analysis in AFL is giving value to those stats. Compare it to the big american sports where it's pretty simple: A running back is judged on metres gained, a point guard on field goal %, a baseballer on batting average. The easy and most common stat for AFL midfielders is disposals but that doesn't tell us much: Was it effective? How far did it go? Handball or kick? Was it an attacking kick i50 or crossing out wide? While the numbers can give us a rough idea of how good a player is, they only go so far. On top of that, there's 18 players out there with more varied roles/positions than what you'd see in American sports. It's hard to compare an inside mid to an outside mid and then you need to consider how much time they spend around the ball compared to time forward/behind the ball.

I completely agree with this. Indeed, that is why the AFL rankings try to account for things like where a posession was gained, how far it went, the value of where it was directed, etc etc. That system does seem to throw up some anomalies unfortunately, though maybe it is the general football community that is undervaluing the likes of Danger, Rioli and Harvey...

 

[Liverpower]

Or to rephrase the above>
"Even though my team is doing really well both on and off the field right now, I'm so miserably absorbed by the AFC that i can't help but put up a bit of BS attacking a player that seems like a pretty decent bloke and a hard at it footballer"

Not absorbed, it was just an interesting question to look at. The numbers I put together can just as easily be used to compare Ablett and Swan, or Watson and Selwood...

 

[Damon_3388]

Consistency year to year? What are you on about? Players assume different roles depending on the year, and injuries. His coach may has asked him to play more of an outside role one year versus the next. What matters is consistency game to game.

Switch of clubs aside, has Judd's role really changed that much from year to year? His tackling numbers yo-yo up and down, week to week and year to year. It's hardly an outlandish call to say that they're A) Inconsistent, and B) He's never consistently been a prolific tackler.

What we do know is that Dangerfield has never been a big tackler. Ever. The comparison you made is ridiculous.

I think you're taking what was originally an off-the-cuff comment a little too seriously. However, I don't think comparing two burst players who've never been great defensively and have both been known to have iffy kicking is ridiculous at all.

 

[Sherrinator]

Some players make contributions that aren't evident in standard stats analysis.

I would be interested to know per game how many metres a player gains to advantage ie. Dangerfield receives the ball, breaks a tackle runs 35 metres and passes the ball 30 for a mark. That's 65metres there. If a players kicks a 60m bomb for a goal or a mark, that is 60m to advantage.

 

[Tas]

What I have done is compared Dangerfield with many of the top midfielders in the league over the past four years across a range of key metrics. The other players are Ablett, Swan, Pendlebury, Griffin, Johnson, Selwood, Mitchell, Watson, Beams, Cotchin, Kennedy and Sloane.

I see the problem.

Compare him to Boak and Hartlett.

 

[StonesThrow]

I'm not sure of this bit I think his tackling numbers are way up this year. Certainly see him making more tackles anyway. Maybe the OP had a word to him in the off season.

To be fair, while being a good tackler, he certainly needed to up his defensive game and make more of them.

Kicking is not his strength and I think it will always be inconsistent unfortunately.

His hard ball gets are the most underrated aspect to his game. Not just the numbers but the way he gets them. Gets the footy to a teammate when he just didn't have the right to. A lot of punters just focus on his burst running and long goals as his number 1 strength.


Game time up forward definitely skews any stats comparing him to other top line mids. He is up forward for a frustrating amount of time as a supporter.

There are definitely other mids ahead of him. Would hate to lose him though. He's my favourite player and imo the most fearless in the comp, which also goes understated.

 

[SydCrow]

I wonder why the official AFL player rankings, which are based on statistics, have him at #3.

 

[Liverpower]

I wonder why the official AFL player rankings, which are based on statistics, have him at #3.

The explanation for that would have to be that when the AFL's researchers look more closely at where Dangerfield gets his posessions, and the quality of those posessions, they are found to be more valuable than my very basic comparison shows. Put differently, their system puts a low weighting on things like tackles (it has a broader set of metrics for identifying pressure acts) and finds his posessions more effective than otherwise seems to be the case.

 

[Liverpower]

I see the problem.

Compare him to Boak and Hartlett.

It is fair to say that statistically he was superior to any of Ports mids in 2012 and 2013. Whether that holds this year remains to be seen

 

[Liverpower]

I'm not sure of this bit I think his tackling numbers are way up this year. Certainly see him making more tackles anyway. Maybe the OP had a word to him in the off season.

To be fair, while being a good tackler, he certainly needed to up his defensive game and make more of them.

Kicking is not his strength and I think it will always be inconsistent unfortunately.

His hard ball gets are the most underrated aspect to his game. Not just the numbers but the way he gets them. Gets the footy to a teammate when he just didn't have the right to. A lot of punters just focus on his burst running and long goals as his number 1 strength.


Game time up forward definitely skews any stats comparing him to other top line mids. He is up forward for a frustrating amount of time as a supporter.

There are definitely other mids ahead of him. Would hate to lose him though. He's my favourite player and imo the most fearless in the comp, which also goes understated.

His relative tackling stats are up this year so far - to 4.2 per game - however, that is still more than two tackles a game behind Ablett, Pendlebury, Beams and Sloane, and also well behind Johnson, Selwood and even Watson.

Would definitely agree that any team would and should want him. One thing that comes through is that no other player has a statistical profile quite like him. That suggests that he would add something unique to all teams...

 

[Tas]

It is fair to say that statistically he was superior to any of Ports mids in 2012 and 2013. Whether that holds this year remains to be seen

How long has Sando been coaching? Because the dill before him played him too much up forward, it is going to be hard to generate core midfielder numbers when you are playing too much up forward.

I haven't seen a lot of Adelaide games this year to date, is he playing much up forward?

 

[DangerSloane]

Something to think about with Dangerfield is he often tries to 'steal' the ball off players with brute strength rather than tackling them.

You think this would be a rare thing, but its actually very common.

 

[DangerSloane]

How long has Sando been coaching? Because the dill before him played him too much up forward, it is going to be hard to generate core midfielder numbers when you are playing too much up forward.

I haven't seen a lot of Adelaide games this year to date, is he playing much up forward?

A little to rest and because hes been a bit battered and bruised trying to avoid the tag.

Nowhere near as much as old mate Craigy though...but he rightfully said that Danger didnt have the tank back then.

He's only been a pure midfielder for really the last 2 seasons ('12-'13)

 

[PowerForGood]

I am amazed that no one has ever out together a full +/- analysis aka basketball. Doesnt matter where you play in the field then.Someone has all the interchange times for every game, the player/time for each disposal, and the time of every score of every game, so is possible with the right algoritm and enough processing power.

Happy to do it if someone had the stats (unlikely Champion Data would give it up any time soon)



Sent from Alberton Oval, the centre of the football universe

 

[Champion_Data]

I am amazed that no one has ever out together a full +/- analysis aka basketball. Doesnt matter where you play in the field then.Someone has all the interchange times for every game, the player/time for each disposal, and the time of every score of every game, so is possible with the right algoritm and enough processing power.

Happy to do it if someone had the stats (unlikely Champion Data would give it up any time soon)

The stat has some merit in theory but across 18 players the effects of having someone on/off the ground is significantly reduced compared to five for basketball.

It has been done and it has been shown to clubs without much interest. Consider the following for a reason why: By this measure Gary Ablett was the worst midfielder in the competition last year. Gold Coast's points per 100 minutes with Ablett on the field was 71.3 for, 79.7 against for a margin of -8.3 points. With him on the bench, it was 97.0 for, 57.4 against for a margin of -25.7 points. His +/- differential of -48.0 points per 100 minutes was the worst of the 150 midfielders who played 10 or more games last season an over 20% worse than any other player (next worst was -39.2 points/100 minutes). This year after six games he is coincidentally running at exactly the same differential as last season (On: 67.9 v 75.3. Off: 94.0 v 53.5 for -48.0) - the eighth-worst of the 123 midfielders with four or more games.

Others in the bottom-20 (all players, not just mids) this year who you may be surprised to see there: Scott Selwood, Michael Johnson, Jared Polec, Sam Mayes, Jordan Lewis, Mathew Stokes, Cale Hooker, Brent Macaffer.