Demons are the team to beat in 2019: Champion Data

[The Chazz 4,077]

58 minutes ago, jnrmac said:

It was Coll as the #1 midfield as well a Melb as the #18th best midfield that had everyone scoffing.

Just like many clubs will be scoffing when they see that CD has Melbourne with the best midfield.

 

[Demon trucker 1,786]

2 hours ago, DubDee said:

You lost me at Adelaide 2nd and West Coast 11th

That's not true, it is wrong and right, Adelaide list is the second best, the biggest difference between the crows, 17 and 18 seasons was injuries, they had hardly any injuries in 17 made grand final, in 17 their list was health, they won most of their last 7 games in 2018,once their list got fit, and if we lost to them in Adelaide they would have had 13 wins in 2018.  with west coast they are way off the mark.. don't be surprised if the eagles drop off next year, they are a old and mature, side who had been in the premiership window for the last 4 years. Most of their players played in the 2015 GF, they did not come from no where.

[Gator 16,921]

9 minutes ago, rjay said:

Maybe opinion is the wrong word, maybe not..

My point is that they chose the stats and the weighting of those stats.

It's not an exact science. Someone else may find a different set of figures and weighting is more representative.

They sometimes overreach with their conclusions but it makes for an interesting debate and they are obviously refining the process each year.

I think your original post 'pro' on the rolling 40 games is a case in point, pretty sure that's how they did it in the past but have moved on and refined it.

This is the point I was getting at...not everything is based on the raw numbers. The stats they chose and weighting are generated by their experts not raw numbers. It's still an opinion based business, in their case what they think is most important in the outcome of games or positions played.

They (CD) do put out their conclusions and use the media to push their profile. The media presentation is usually their (CD) representation of the data in press releases and they have their people talking about it on as many shows as possible. Good for business.

Our game is a very difficult one to boil down to just a set of raw data...much easier for other games like Cricket & Baseball.

Whilst like you I don't agree with the post you were originally quoting 'Brenno', that champion data is garbage, I think they do bring some really interesting information to the table and create some good debate/arguments...I don't totally buy into what they have to say either, raw data or not it's not above reproach.

The topic doesn't overly interest me, but having heard a CD rep on SEN this morning it was clear that none of this is subjective opinion.

It's 100% raw stats.

As to the poignancy of the stats or how they rate each stat I wouldn't know.  But they've been doing stats for over 20 years, so I suspect their models would be reasonably relevant.

[ManDee 7,093]

1 hour ago, ProDee said:

I don't think this is correct.

CD said today on SEN that it is entirely stats based.

They may give more weight to some stats over others, but none of their assessments are linked to "opinion".

The problem with weighting is it can be opinion. Are handballs worth more than effective hitouts? Is an effective 50 metre kick across ground worth more than a 20 metre kick into the forward line? Is a mark in the defensive 50 in junk time worth as much as a mark on the wing? And so it goes.

[DubDee 21,400]

1 hour ago, Watson11 said:

Based on the stats West Coast's midfield was middle of the road for most of 2018.  They had a 3 week purple patch that just happened to co-incide with the finals, helped hugely by Gaff replaced by Redden who averaged around 15 CPs through the finals.

Adelaide will be top 4 this year.

they were first or second on the ladder all year long. and played without many of their best players throughout the season and finals.  Gaff, Nic Nat, Darling, Kennedy

CD analysis is garbage

[Gator 16,921]

15 minutes ago, DubDee said:

they were first or second on the ladder all year long. and played without many of their best players throughout the season and finals.  Gaff, Nic Nat, Darling, Kennedy

CD analysis is garbage

It's not an "analysis", it's stats.

For example, West Coast ranked 18th for ground-ball gets. 

Stats where they weren't great during the H&A they excelled at in finals.  They weren't a great contested ball winning team whereas Melbourne were no.1 in the competition, but who won the contested stats in the prelim ?

Sheed had been in and out of the side during the year, but was great in the finals.  Redden had a great last 6 weeks.

West Coast were 18th for ground-ball stats, 11th for contested possessions, 11th for clearances, and 15th for tackles.

It is what it is.  It's just stats.

[DubDee 21,400]

1 hour ago, ProDee said:

It's not an "analysis", it's stats.

For example, West Coast ranked 18th for ground-ball gets. 

Stats where they weren't great during the H&A they excelled at in finals.  They weren't a great contested ball winning team whereas Melbourne were no.1 in the competition, but who won the contested stats in the prelim ?

Sheed had been in and out of the side during the year, but was great in the finals.  Redden had a great last 6 weeks.

West Coast were 18th for ground-ball stats, 11th for contested possessions, 11th for clearances, and 15th for tackles.

It is what it is.  It's just stats.

Alright, I'll change the phrasing:

Ranking sides based on a combination of stats does not indicate who will win the flag.  how can we be  the 2nd best defence? even us supporters know this was our weakness all year.  

it's pointless imo

[jnrmac 19,336]

2017 CD rankings FWIW and where they finished on the ladder at H&A

1. GWS (7)
2. Sydney (6)
3. West Coast (2)
4. Hawthorn (4)
5. Western Bulldogs (13)
6. Adelaide (12)
7. Port Adelaide (10)
8. Collingwood (3)
9. Geelong (8)
10. St Kilda (16)
11. Melbourne (5)
12. Richmond (1)
13. North Melbourne (9)
14. Fremantle (14)
15. Essendon (11)
16. Gold Coast (17)
17. Carlton (18)
18. Brisbane (15)

[Gator 16,921]

10 minutes ago, DubDee said:

Alright, I'll change the phrasing:

Ranking sides based on a combination of stats does not indicate who will win the flag. 

Where did you read that this was a flag prediction ranking ?

[Chook 14,616]

This is the best day of my life. It's even better than the day I read we were the best Victorian team in the league…back in 2007, when I was young and happy and didn't know what I didn't know.

In the words of Bob Seger, "I wish I didn't know now what I didn't know then."

Edited November 29, 2018 by Chook

[FlashInThePan 1,893]

1 hour ago, DubDee said:

Alright, I'll change the phrasing:

Ranking sides based on a combination of stats does not indicate who will win the flag.  how can we be  the 2nd best defence? even us supporters know this was our weakness all year.  

it's pointless imo

They are not ranking the defenders we had all year. Lever was injured most of the year and May was at the gold coast. They are ranking our best fully fit defense we would have in 2019.

[Watson11 1,934]

1 hour ago, DubDee said:

Alright, I'll change the phrasing:

Ranking sides based on a combination of stats does not indicate who will win the flag.  how can we be  the 2nd best defence? even us supporters know this was our weakness all year.  

it's pointless imo

One good thing about stats is they don’t suffer from confirmation bias.  Not all us supporters know our defence was our weakness.  Some of us attribute our habit of leaking goals to our forward 50 and midfield pressure going missing at times.

[old55 21,749]

Clubs pay many $000s for CDs standard stats package and many more for custom packages.  I don't think they regard them as useless.

What seems to missing here is the realisation that CD correlate stats with game outcomes to determine the important stats. It's an emerging and imperfect science but it's not "opinion".

[4_Kent_Watts 850]

Well I guess my SuperCoach team will be filled by only Red and Blue jumpers this year. Bring it on.

[DubDee 21,400]

5 hours ago, ProDee said:

Where did you read that this was a flag prediction ranking ?

sorry - 'team to beat' was the wording used

[DubDee 21,400]

4 hours ago, FlashInThePan said:

They are not ranking the defenders we had all year. Lever was injured most of the year and May was at the gold coast. They are ranking our best fully fit defense we would have in 2019.

righto, ive got the wrong end of the stick.  should properly read the article next time

[Males 578]

4 hours ago, Fifty-5 said:

Clubs pay many $000s for CDs standard stats package and many more for custom packages.  I don't think they regard them as useless.

What seems to missing here is the realisation that CD correlate stats with game outcomes to determine the important stats. It's an emerging and imperfect science but it's not "opinion".

Not only that, CD main man Glenn Luff was poached by North Melbourne to be their stats guru this year. They certainly rate their stats.

[Guest]

As Jerry Seinfeld says at the end of this clip...

 

[nosoupforme 3,085]

Champion Data so called have the Eagles ranked at no. 11  and with a Premiership to boot. However they now strengthen their team with Gaff. Collingwood    no 5  with the Data add Beams and Elliott if fit and so on. IMO the hardest sides to beat. Richmond ,Essendon  and ,Hawthorn no 3, 4 and 9 ranked side look quite strong. Interesting times in 2019.

In the meantime the Dees are the no 1 ranked side. l will take that and hope we defend it resolutely.   The inclusions of May and Lever,  ,barring any major injuries will help the cause.

[dee-tox 4,830]

Bill Clinton might be tempted to say what many of us believe: It's the game plan stupid. Stats don't tell even half the story.

[La Dee-vina Comedia 16,534]

How unhygienic is it for Champion Data to be drinking our bathwater?

It seems to me that Champion Data's statements are based on raw stats processed by an algorithm. There must be assumptions incorporated into the algorithm and those assumptions are created by people. Hence, it is not entirely incorrect to refer to CD "analysing" the stats and providing its "opinion". Seems to me, though, that the moment organisations state that they've used an "algorithm" to do something, the belief given to whatever follows seems to increase by about 33%. (Note: that last figure is not created by an algorithm so please treat it with caution.)

[Gator 16,921]

11 hours ago, DubDee said:

sorry - 'team to beat' was the wording used

That's a sub editor's headline.  It's a journo's interpretation.

It's not a comment from CD.

[tiers 2,785]

It seems to me that the one stat that is missing is "value per possession". Of course this is so subjective to be impossible to measure but given that weagles stats in 2018 were not elite then perhaps there is something else out there.

When I see Tracca receive the ball in a pack and in the mayhem turn around and off one step kick a goal tells me that possessions and score involvements are not enough. Or when I see Tracca receive in midfield, spot and deliver perfectly to Spargo who in a blink handballs to TMac for an easy goal, raw stats and algorithms become meaningless.

It's all about that immeasurable quality - footy smarts. And we are starting to accumulate it is spades.

2020

Go dees.

[Gator 16,921]

The Eagles stats were elite when it mattered - finals.

The competition is so close with no standout teams.  If you make finals and perform at the right time you can walk away with the chocolates Footscray-style-2016.

[jnrmac 19,336]

16 hours ago, Watson11 said:

One good thing about stats is they don’t suffer from confirmation bias.  Not all us supporters know our defence was our weakness.  Some of us attribute our habit of leaking goals to our forward 50 and midfield pressure going missing at times.

I am shocked to read this. According to 'Land we have the best defence over the last 2 years by a country mile.