Champion Data’s List & Player Ratings - An Accurate Tool or Total Fabrication ??

[Boxhead_31]

Gets e even better when you see the players CD rate "Elite"

Sorry Hawk fans Tex Walker, Connor McKenna, Huge Greenfield and Majak Daw are Elite talent and Tom Mitchell isn't

 

[Monkey King]

I take this is a huge grain of salt

I think you want one of these ... yours is just a big pile of little grains of salt.

 

[TheWoodenSlug]

Gets e even better when you see the players CD rate "Elite"

Sorry Hawk fans Tex Walker, Connor McKenna, Huge Greenfield and Majak Daw are Elite talent and Tom Mitchell isn't

I don't even know who a bunch of those guys are. T. Walker elite, but Hawkins, Kennedy and Riewoldt aren't? Go home CD, you're drunk.

 

[Red_man2008]

Love champion data statistics but disagree with the elite category they come up with. They use 5% of players from the entire afl in that position and classify as it elite. BUT how does ward make that list. He didn’t even play a game. Daw. Did he play a game. I understand their historical data may come into it but then Docherty should still be an elite HBF. It’s confusing their cut offs. I say you need to play a minimum of 10 games in 2019 to classify.

How can someone like stack sit in front of Walsh unless he’s got the role off a halfback which most sides rotate players through. Whenever one player spends the majority of their time off a wing they seem to get onto the elite list. See Robinson and recently Polec. Why do they have a mid-forward position just for Dustin Martin. Did they invent that category just to avoid criticism of their data. I think so. Why isn’t there another person??

 

[HairyO]

Who did they put ahead of Sicily for intercept and D50 rebound ?

Lots of mocking of Henderson but his last 2 years were as good as any other wing in the comp.

 

[TheKITC]

Any list that claims to rate 'elite' players and has the likes of Hugh Greenwood, Mitch Robinson, Majak Daw, Anthony McDonald-Tipungwuti, Matt Suckling, Taylor Walker etc. on it, but has no Scott Pendlebury, Tom Mitchell, Tim Kelly etc. speaks for itself. Clearly flawed in it's parameters considered and not worth taking seriously.

 

[KermitJagger]

Who did they put ahead of Sicily for intercept and D50 rebound ?

Lots of mocking of Henderson but his last 2 years were as good as any other wing in the comp.

Henderson is elite because he's elite for his position, Champion Data for some inane reason have very few players classed as genuine wingmen. It's the same reason that Paul Seedsman and Trent Dumont were classed as elite for this year.

 

[Engimal v3]

Any list that claims to rate 'elite' players and has the likes of Hugh Greenwood, Mitch Robinson, Majak Daw, Anthony McDonald-Tipungwuti, Matt Suckling, Taylor Walker etc. on it, but has no Scott Pendlebury, Tom Mitchell, Tim Kelly etc. speaks for itself. Clearly flawed in it's parameters considered and not worth taking seriously.

I believe CD's definition of 'elite' means by position. So AMT is probably elite for a small forward, but not a player you'd consider to be elite overall.

 

[HairyO]

Henderson is elite because he's elite for his position, Champion Data for some inane reason have very few players classed as genuine wingmen. It's the same reason that Paul Seedsman and Trent Dumont were classed as elite for this year.

Which I think is fair. If they werent classing players based on position then you would have a team of the best 22 midfielders which would be absolutely flogged by the best 16 other positions and the 23rd through 28th best midfielders.

 

[Underarm]

Which I think is fair. If they werent classing players based on position then you would have a team of the best 22 midfielders which would be absolutely flogged by the best 16 other positions and the 23rd through 28th best midfielders.

Champion Data do such a poor job explaining this though, and the Media run with the numbers without explaining them.

Their 'elite' rankings over represent Ruckman (7/51 ranked elite) and under represent midfielders (16/51). They're so hard to make sense of that I dont put much stock in them when they are used.

 

[DamoESP]

The fact Taylor Walker is listed as an "elite" player just goes to show everything you need to know about the list.

 

[nineteen eighty]

Generally speaking, CD have two forms of data:

1/ Raw output (I50's, contested possessions, clearances etc...)
2/ Data that is manipulated to tell a story (such as this garbage analysis).

I enjoy the raw data. It tells a story about players / teams. Make of it what you like.

The second form...yeah nah, not for me.

 

[HairyO]

Generally speaking, CD have two forms of data:

1/ Raw output (I50's, contested possessions, clearances etc...)
2/ Data that is manipulated to tell a story (such as this garbage analysis).

I enjoy the raw data. It tells a story about players / teams. Make of it what you like.

The second form...yeah nah, not for me.

This is the thing. The clubs get the raw data and use it their own way. And they seem happy with what they get.

So its not so much the data at fault, but how it is used.

Its been obvious for a very long time that CD use it badly. Or they just enjoy trolling with clickbait way too much.

 

[Rimmer]

Champion Data are quite good they are just usually a year behind. Melbourne were close to the best in 2018 and Adelaide were likely the best in 2017. There does not seem to be a way to put a stats value on running like the wind, chasing and just pushing the ball forward. So the stats make interesting reading but are meaningless when trying to accurately assess the value of a player to his team. The guy who gets 7 possessions with 7 tackles may be more worthwhile to a certain team than another 30 touch a game player. The right piece in the right puzzle.

 

[Adelaide Hawk]

CD ratings are pile of steaming horse crap.

 

[TheKITC]

I believe CD's definition of 'elite' means by position. So AMT is probably elite for a small forward, but not a player you'd consider to be elite overall.

Is still a massive stretch to suggest he is elite in this category even, but I guess that's pretty subjective as well.

If 18 teams have two small forwards on any given day, which I think is reasonable... elite for mine would be the top 5-6 of a list of 36, let's call it 40 factoring teams like WCE for example. Many I would have ahead of AMT.

Charlie Cameron, Tom Papley, Michael Walters, Toby Greene, Gary Ablett, Robbie Gray and Jade Gresham are comfortably better. I reckon Connor Rozee, Linc McCarthy, Liam Ryan and Will Hayward would have a fair claim to being more if not equally as effective and consistent as AMT, who at times is a bit of a passenger.

Just a s**t rating system...

 

[BigVic]

Don’t trust CD player ratings at all. They’re a myth only

 

[Final Siren]

Champion Data do such a poor job explaining this though, and the Media run with the numbers without explaining them.

This is a big part of it. CD don't see themselves as a public-facing organization. They produce some truly excellent analysis - and, of course, generate most of the fundamental stats in the first place - but by the time we see it, it's passed through several sets of hands, including one belonging to a journo or exec with an eye on headline clicks.

Stats usually address a very specific question, and when we don't know exactly what the question was, the answer may not make a whole lot of sense.

 

[nineteen eighty]

This is the thing. The clubs get the raw data and use it their own way. And they seem happy with what they get.

So its not so much the data at fault, but how it is used.

Its been obvious for a very long time that CD use it badly. Or they just enjoy trolling with clickbait way too much.

It's content that they're producing. Problem is that they seem blinded to some of their garbage content. As you said, the raw data is fine. Very informative and very much appreciated.

 

[Hoops]

That Chris Scott's a good coach, taking a team ranked 7th to 3rd

 

[Colonial]

How do the Pies, Dogs and Tigers keep there talent under the cap?

Something dodgy happening here.

 

[doppleganger]

Champion Data are quite good they are just usually a year behind. Melbourne were close to the best in 2018 and Adelaide were likely the best in 2017.

That is because they are just taking the descriptive stats they capture based on previous games and seasons, and using that as the basis for his is ‘best’ for the upcoming season.

They are great at capturing descriptive stats, but as the super funds always disclaim - Past performance is no guarantee of future results.

The player ratings are a useful tool to help actually compare players, but again need to understand the weightings if respective stats properly for any meaningful comparison to be made.

The problem is too many people don’t actually understand the stats they try and use to support an argument, or bother to understand the CD definitions, and hence they are often used poorly by punters.

 

[KermitJagger]

That is because they are just taking the descriptive stats they capture based on previous games and seasons, and using that as the basis for his is ‘best’ for the upcoming season.

They are great at capturing descriptive stats, but as the super funds always disclaim - Past performance is no guarantee of future results.

The player ratings are a useful tool to help actually compare players, but again need to understand the weightings if respective stats properly for any meaningful comparison to be made.

The problem is too many people don’t actually understand the stats they try and use to support an argument, or bother to understand the CD definitions, and hence they are often used poorly by punters.

Wittgenstein's ruler - without confidence in the reliability of a ruler, if you use a ruler to measure a table, you may also be using the table to measure the ruler.

That's pretty much what it is, using the table to measure the ruler.

If you wanted a better metric to judge whether a player is elite or not, it should probably be averaged out over a few years. Then have a pyramid scale based on age, where you might get a relative bump if you're younger than 28-29 (peak for most players) and then a shift downwards for every year following.

 

[Sweet Jesus]

More weirdness...

Champion Data ranks all 18 AFL teams by each position

Champion Data has ranked every AFL team, broken up by their defenders, forwards, midfielders and overall.

www.sen.com.au

 

[rumply]

More weirdness...

Champion Data ranks all 18 AFL teams by each position

Champion Data has ranked every AFL team, broken up by their defenders, forwards, midfielders and overall.

www.sen.com.au

Overall rankings actually makes sense to me as far as Essendon is concerned but GWS's midfield below Essendon's midfield? Yeah that is a bit weird...