Herkes tarafından bilinerek sevilen site olan 1xbet canlı adresi sizlere büyük avantajlar ile farklı bahis imkanları sunmaktadır. Bilindik bir firma olması nedeni ile her defasında yeni bir 1xbet güncel adrese taşınıyor. Paylaşılan adreslerden sizlere en uygun 1xbet türkiye giriş güncel adresine kolaylıkla hemen ulaşabilirsiniz. Sizlerde kolaylıkla her cihazınızda aktif olan 1xbet mobile ile bahis yaparak, üyelik oluşumunu halledebilirsiniz. Büyük promosyonlardan yararlanarak üyelik açmak için 1xbet live adresini kullana bilirsiniz. Üyelik oluşturduktan sonra kolaylıkla yatırım yapmak için mobil ödeme bahis kabul gördüğünü anlayabilirsiniz. Hiç bir yerde olmayan canlı bahis özelliklerini sizlerde hemen kullanın. Aktif bir şekilde işlem yapan canlı bahis sitesi editörler tarafından özenle araştırılarak seçilmiştir. Ülkemizde resmi yayın yapmayan sitelerin çoğu kaçak bahis adı altında görev yapmaktadır. İnternetten yayın yapan kaçak bahis siteleri kullanıcılarına yüksek oranlar sunan bir adrestir. Hemen sizlerde casino oyunun farkına ve eğlencesine varmak için kayıt oluşturun.
Forum
Index
Docker Discussions
Champion Data
  • Page:
  • 1
  • 2

TOPIC: Champion Data

shane Champion Data 5 years 1 month ago #1

shane
I was reading this article about what Nick Riewoldt reckons. I don't know why I was reading it.

www.foxsports.com.au/afl/afl-2019-nick-r...220c16553d512ca9dd07

Of all AFL teams so far this season, the Bombers have undoubtedly attracted the most criticism after poor losses to the GWS Giants and St Kilda, with Friday night’s clash against fellow strugglers Melbourne set to be a grand talking point.

The losses came after many pundits pre-season predicted John Worsfold’s men would return to the finals — predictions that were backed up by the fact Champion Data rated Essendon’s list as the fourth-best in the competition.

Essendon are rubbish, which, while they could have gone the other way, it can't really be that big a surprise for a team that went through a quite a bit of an upheaval. I'm not sure the idea that pundits got something wrong should be held against the club.

But more to the point, Champion Data. Every year they seem to get a stronger profile in the AFL system, I imagine they're getting a decent amount of money for collecting the statistics but they seem to have found a way to get a lot more money by providing analysis of these statistics.

The thing is, they don't seem to be particularly good at that. Their rankings don't seem to be any better than the average pundit, much of the rankings they produce don't match with reality. I know that's often said about data versus perception but this isn't data, it's someone's modelling of data and I tend to think the assumptions they feed into their models is questionable.

I don't know if this data is used for anything other than entertaining the public but I'd be worried to think their modelling was being used in decision making.
Login to reply,
goodie said You Beaut

rogerrocks Champion Data 5 years 1 month ago #2

rogerrocks
Like dividing the AFLW into two *even* conferences?
Login to reply,
teasea said You Beaut

shane Champion Data 5 years 1 month ago #3

shane
Exactly like that.

I don't know if that happened but it would fit my modelling.
Login to reply,
blockerhall, pollyanna said You Beaut

Jooter Champion Data 5 years 1 month ago #4

Jooter
Funny you should bring this up, Shane. I just got into this fascinating podcast:
www.afl.com.au/news/2019-03-27/listen-ne...tes-footballs-trends

While it is incredibly informative, has discussions that are data-driven, the insights suggest that (all) clubs use this kind of analysis to help make decisions, but that in-game changes in trends may have a smallish role to play in actually changing the outcome of a game. Statisticians and keepers of the data feed info to coaches, who generally have already seen or had a feel for the trend that is emerging. That is, the data is used to 'back up' what they see.

A few other points I got out of the first two episodes:
- each team trains its game plan over pre-season and gets feedback from the first six-or so games before actual tweaking may occur. In essence, most teams will 'play their own game' rather than just react to what is going on
- sometimes there's a clear disconnect between what the coach says to do and what happens in the heat of the moment with a player (for me, this came down to the last play of the Freo-Suns game: Hill 'bombed' it in to a pack, whereas a more prudent kick would have been to a target; in fact, you can see Taberner signal with his hands to keep the ball low into the forward line, whereas Sam Collins famously marked the incoming ball)
- data is everything these days, and the AFL leads comps around the world in using the data. Is this a good thing? I don't know, but to me, it represents that extra 1-3% improvement clubs are seeking over each other in a brutal comp like the AFL.
- Certainly, this focus on data is a _clear_ improvement over the casual fan who just says a team loses because they're not trying hard enough. I hear that enough at the footy to know it's complete nonsense.
- Finally, woe-betide the coach who doesn't use this type of data because they are the ones who will not only consistently lose, but won't have the back up of a data driven approach to explain the losses.

So, is this data-driven trend a good thing? It's kind of a moot point, because all clubs are doing it, and without it, then that's at least something a Board can point to - that coaches aren't using this approach - to explain why a team consistently loses.
Login to reply,
shane said You Beaut

shane Champion Data 5 years 1 month ago #5

shane
I have no problem with being data driven. Although I think the game would be more pure if those involved in the contest didn't have access to live data during the game.

My issue is how they interpret the data. The consumer grade analysis that we seed from Champion Data, combined with how you hear former players and coaches talking about data doesn't fill me with confidence. The experts who have been through the system and explain how the coaches look at data have sketchy ideas about causality.

But more to the point, f the AFL are making decisions on fixtures, on rule changes, on anything really, and the interpretation of the data they use says that Essendon will be a top 4 team, and Essendon are not a top 4 team, then they're paying a company a lot of money to feed them useless information and making bad decisions because of it.
Login to reply,
Jooter said You Beaut

The_Yeti Champion Data 5 years 1 month ago #6

The_Yeti
To be fair to the AFL, Shane, if the AFL didn't make bad decisions, the game wouldn't be where it is today.
Egurls Suck!
Login to reply,
shane, Raglan Matt said You Beaut

shane Champion Data 5 years 1 month ago #7

shane
But if they didn't pay for the expert advice, they'd still make bad decisions but they'd have slightly more money, so maybe I could go to the football and afford to drink water that didn't come from the toilet.
Login to reply,
The_Yeti said You Beaut

Raglan Matt Champion Data 5 years 1 month ago #8

Raglan Matt
Mates of mates making money, it all adds up when you look at it like that.
Login to reply,

Wedge Antilles Champion Data 5 years 1 month ago #9

Wedge Antilles
"data is everything these days, and the AFL leads comps around the world in using the data"

Jooter, the AFL has licensed the data to one company, Champion Data. We therefore have one source for this analysis.

I would suggest the AFL trails most other major sporting competitions around the world when it comes to data.

The NBA and Major League Baseball are the obvious examples. They leave the AFL in the dirt when it comes to statistical analysis for the simple fact that the data is open source, so you have multiple Champion Data's with competing ideas.

In fact, you have individuals who can analyse the data and come up with competing theories hence Moneyball et al.

You won't see this on the AFL website - nbasense.com/nba-api/
Login to reply,
shane, pollyanna said You Beaut

Walter the baker Champion Data 5 years 1 month ago #10

Walter the baker
While stating the obvious, Champion Data are simply using past data and then extrapolating into the future as though nothing else changes. Unfortunately, we all know that nothing will be the same when two teams run out next week. Most significantly 80% (s figure I plucked out of the air) of the game is played above the shoulders - as highlighted by the two Bombers players who ignored the ball last week because they were thinking of something else. Until Champion Data can work out how a player is going to be feeling during a game their predictions will always be educated guesswork.
Login to reply,
rogerrocks said You Beaut

The_Yeti Champion Data 5 years 1 month ago #11

The_Yeti
You would have to question the value of extrapolating data into, as has been pointed out, a changed environment. You would have to question further it's value in defining major management decisions, particularly when any analysis can take any set of figures and produce a desired result.

One example of this:
The Japanese eat very little fat and suffer fewer heart attacks than the British or Americans.
The French eat a lot of fat and also suffer fewer heart attacks than the British or Americans.
The Japanese drink very little red wine and suffer fewer heart attacks than the British or Americans.
The Italians drink excessive amounts of red wine and also suffer fewer heart attacks than the British or Americans.
The Germans drink a lot of beer and eat lots of sausages and fats and suffer fewer heart attacks than the British or Americans.

Conclusion: Eat and drink what you like. Speaking English is what kills you.

Its a logical conclusion based on the facts that were used to derive it but unless you take into account each and every component, you won't get an entirely accurate extrapolation.

Champion Data is a very limited set of measurements, recorded with almost no context and no valid comparisons between games or seasons. At best, it 'might' be indicative but equally, it 'might' be misleading. What is clear is that it is not a viable management tool.
Egurls Suck!
Login to reply,
Raglan Matt said You Beaut

shane Champion Data 5 years 1 month ago #12

shane
That's down to interpretation of the data though. From what I remember, someone made the conclusion that the French suffered fewer heart attacks because of their diet when it was actually just the way the cause of death was reported in France. The assumptions used to draw the conclusions were wrong.

I bet if the right people got hold of the data they could build much better models than produce more informative results than what Champion Data are pumping out. If not, then the AFL needs to stop buying the information.
Login to reply,
The_Yeti said You Beaut

Raglan Matt Champion Data 5 years 1 month ago #13

Raglan Matt
You are a genius, Shane;
The assumptions used to draw the conclusions were wrong.

Pity the AFL and media don't hire you as their analyst.
Login to reply,

rogerrocks Champion Data 5 years 1 month ago #14

rogerrocks
The trouble with data is that it is a chicken and egg thing. Did the team win because they had better disposal? Or were their disposals better because they had run the opposition ragged and were no longer kicking to a contest, but to a bloke on his own?
If a particular stat is shown to have a strong correlation with winning, then presumably most coaches either try and improve their own performance on that stat, or try to nullify the oppositions performance on that stat, and suddenly that stat doesn't correlate much with winning any more.
So it will come down a coaches interpretation of what is working and not working, maybe helped by the stats a bit.
(I trust that no one will take this as an invitation to kick any particular coach)
Login to reply,
  • Page:
  • 1
  • 2
Forum
Index
Docker Discussions
Champion Data