| This alternative table garbage 19:55 - Sep 3 with 13412 views | bluenile |
Don't know if anyone else read that alternative table article? Apparently based on where they thought every team was expected to be, not where they actually are, after 6 games. And I couldn't bring myself to look at the bit about "Swansea, papering over the cracks?" . . . . . . . . Christ on a bike! |  |
| Open the ipod bay doors Hal |
| |  |
| This alternative table garbage on 12:23 - Sep 4 with 1957 views | Swanjaxs |
Talkshyt should be banned anyway [Post edited 4 Sep 2019 12:25]
|  |
| |
| This alternative table garbage on 12:31 - Sep 4 with 1959 views | Captain_Sham |
| This alternative table garbage on 10:24 - Sep 4 by Professor |
Any statistical analysis or (in the case of xG) predictive mathematical model is only as good as the data you analyse. Football, like a complex biological system, has many variables and confounders which lead to a relatively weak model (its not just us but the Wurzels who are a long way out). A confounder is that Bamford could not hit a bovine rectum with a stringed instrument.
My modelling colleague up in Newcastle always want as much reliable data, preferably longitudinal, as possible to allow decent analysis. This has worked for lab/experimental infection work where we control the vast majority of factors, but was just noise from chicken farms as there were too many variables.
Remember all models are wrong, but some are useful. [Post edited 4 Sep 2019 11:47]
|
Jasper has no understanding of statistics. He just touts them about and thinks they make his opinion valid. He is comical. [Post edited 4 Sep 2019 12:36]
|  |
| |
| This alternative table garbage on 12:42 - Sep 4 with 1941 views | MrSwerve |
xG is a load of rubbish and I cringe whenever I see it quoted anywhere or displayed on TV.
Who cares about quality of chances if your striker can't put it away. There are too many variables, player talent being one, to make it meaningful. That Talksport table proves that. Look at how many teams are in wildly different positions compared to their real life positions.
Just another stat that pundits can waffle about. |  |
| |
| This alternative table garbage on 13:45 - Sep 4 with 1900 views | Groo |
All about quality of chances, nothing about quality of defences to block these. |  |
| Groo does what Groo does best |
| |
| This alternative table garbage on 14:33 - Sep 4 with 1859 views | JackSomething |
| This alternative table garbage on 10:24 - Sep 4 by Professor |
Any statistical analysis or (in the case of xG) predictive mathematical model is only as good as the data you analyse. Football, like a complex biological system, has many variables and confounders which lead to a relatively weak model (its not just us but the Wurzels who are a long way out). A confounder is that Bamford could not hit a bovine rectum with a stringed instrument.
My modelling colleague up in Newcastle always want as much reliable data, preferably longitudinal, as possible to allow decent analysis. This has worked for lab/experimental infection work where we control the vast majority of factors, but was just noise from chicken farms as there were too many variables.
Remember all models are wrong, but some are useful. [Post edited 4 Sep 2019 11:47]
|
Sounds right to me.
Moneyball worked for the Oakland A's. Maybe that was a one-off, a fluke. It helped that baseball was already a ridiculously stats heavy sport.
From my understanding, which basically consists of watching the film, it was also about identifying players that were undervalued elsewhere, but who would be worth more to the A's.
You could argue many football clubs do that already. I doubt many scouts in Britain were recommending the likes of Rangel, Bodde and Scotland. Or would have suggested that journeyman Routledge would find a long-term home at his new club, or that Michu would be a PL sensation based off one season in La Liga.
Perhaps moneyball in football terms just means having a great scouting department and a clear vision for transfers? |  |
| You know, Hobbes, some days even my lucky rocket ship underpants don't help. |
| |
| This alternative table garbage on 14:47 - Sep 4 with 1851 views | Professor |
| This alternative table garbage on 14:33 - Sep 4 by JackSomething |
Sounds right to me.
Moneyball worked for the Oakland A's. Maybe that was a one-off, a fluke. It helped that baseball was already a ridiculously stats heavy sport.
From my understanding, which basically consists of watching the film, it was also about identifying players that were undervalued elsewhere, but who would be worth more to the A's.
You could argue many football clubs do that already. I doubt many scouts in Britain were recommending the likes of Rangel, Bodde and Scotland. Or would have suggested that journeyman Routledge would find a long-term home at his new club, or that Michu would be a PL sensation based off one season in La Liga.
Perhaps moneyball in football terms just means having a great scouting department and a clear vision for transfers? |
I think good analysis can help identify both areas for improvement and potential transfers, but probably no more than 'gut feeling'. People really don't understand statistics beyond mean (or average) and certainly not probability, variation. So we use a lot of Bayesian approaches which look at repeated iteration of the same events with all possible variations-often called Monte Carlo models. They can throw up things (often biased) observation misses. But to work these need good observed data. Baseball and cricket will better lend themselves to this-e.g. Steve Smith is more vulnerable to left-arm spin as the interaction is more bowler/pitcher to batsman rather than football/rugby where there are more interactions.
Statistics are important but football stats are just lists really. |  | | |
| This alternative table garbage on 15:36 - Sep 4 with 1832 views | JackSomething |
| This alternative table garbage on 14:47 - Sep 4 by Professor |
I think good analysis can help identify both areas for improvement and potential transfers, but probably no more than 'gut feeling'. People really don't understand statistics beyond mean (or average) and certainly not probability, variation. So we use a lot of Bayesian approaches which look at repeated iteration of the same events with all possible variations-often called Monte Carlo models. They can throw up things (often biased) observation misses. But to work these need good observed data. Baseball and cricket will better lend themselves to this-e.g. Steve Smith is more vulnerable to left-arm spin as the interaction is more bowler/pitcher to batsman rather than football/rugby where there are more interactions.
Statistics are important but football stats are just lists really. |
Well despite having no clue about Bayesian approaches or Monte Carlo models, that paragraph makes more sense about the use of statistic in sport than most others I have read. | |
| You know, Hobbes, some days even my lucky rocket ship underpants don't help. |
| |
| This alternative table garbage on 18:25 - Sep 4 with 1781 views | Professor |
| This alternative table garbage on 15:36 - Sep 4 by JackSomething |
Well despite having no clue about Bayesian approaches or Monte Carlo models, that paragraph makes more sense about the use of statistic in sport than most others I have read. |
The Monte Carlo is prosaic. It’s like running a roulette wheel thousands of times. If no bias every number should have same frequency. If a number (or anything being measured) comes up more or less frequency it is having a positive
Or negative effect on the model- likely to be important | | | |
Login to get fewer ads
| This alternative table garbage on 19:10 - Sep 4 with 1755 views | DrGonzo |
Stats are for bores with nothing interesting to say. This is xG thing is a whole new level of bollocks. |  | | |
| This alternative table garbage on 19:27 - Sep 4 with 1746 views | MrSwerve |
Here’s a stat - I did a statistics degree and 100% hated it.  | |
| |
| This alternative table garbage on 10:27 - Sep 5 with 1655 views | Badlands |
| This alternative table garbage on 19:10 - Sep 4 by DrGonzo |
Stats are for bores with nothing interesting to say. This is xG thing is a whole new level of bollocks. |
Statistics are vital |  |
| |
| This alternative table garbage on 14:13 - Sep 5 with 1611 views | Phil_S |
| This alternative table garbage on 10:27 - Sep 5 by Badlands |
Statistics are vital |
They are especially the key ones that matter
Won 5 Drawn 1 Lost 0 Position 1st.
I'd happily have had the lowest shots in the division, the least possession, the most chances against and the worst haircuts of all time if we turned the above stats in over a 46 game season
Seems that us doing well on the pitch doesnt resonate with some. |  | | |
| This alternative table garbage on 14:20 - Sep 5 with 1598 views | 34dfgdf54 |
| This alternative table garbage on 10:27 - Sep 5 by Badlands |
Statistics are vital |
Nah.
Maybe for computer games, but this is real life now. |  | | |
| This alternative table garbage on 14:42 - Sep 5 with 1580 views | Uxbridge |
| This alternative table garbage on 19:27 - Sep 4 by MrSwerve |
Here’s a stat - I did a statistics degree and 100% hated it. |
I did part of my degree in it too. I enjoyed it. But then I didn't do my full degree in it 
As anyone who has studied statistics will know, or anyone with an ounce of common sense come to that, they're only ever an aid and the important bit is how you use them, not the bare stats themselves, no matter how clever or whizzy they are.
Not that xG is particularly clever anyway, for all the reasons outlined. Blind adherence to them is as stupid as ignoring them entirely. |  |
| |
| This alternative table garbage on 16:21 - Sep 5 with 1538 views | DrGonzo |
| This alternative table garbage on 10:27 - Sep 5 by Badlands |
Statistics are vital |
Case in point. | | | |
| This alternative table garbage on 17:49 - Sep 5 with 1517 views | Tummer_from_Texas |
| This alternative table garbage on 22:39 - Sep 3 by thornabyswan |
What don't you like.
Being top of the league
3rd round of the league cup
Winning at Leeds with a skillful gutsy display.
The club showing ambition to keep Baston and Ayew.
The bookies having us 4th favourites to win the league they were predicting a bottom half finish before the first game.
Come on Jasper you are on a different planet mate. |
Being wrong appears to be the answer here. |  |
| |
| This alternative table garbage on 18:58 - Sep 5 with 1479 views | perchrockjack |
Tummer.
Serious question,as they say.
When you see your country being ritually abused and insulted,how do you avoid seeing your backside .
We slag off America but,fook me,Trump or not, we love going there.
This is rank hypocrisy and we stink of it.
People from Uk will visit NYC,LA, even Houston without hesitation but baulk at a self drive holiday in Eastern Cape . | |
| |
| This alternative table garbage on 19:03 - Sep 5 with 1479 views | ploppy |
| This alternative table garbage on 14:47 - Sep 4 by Professor |
I think good analysis can help identify both areas for improvement and potential transfers, but probably no more than 'gut feeling'. People really don't understand statistics beyond mean (or average) and certainly not probability, variation. So we use a lot of Bayesian approaches which look at repeated iteration of the same events with all possible variations-often called Monte Carlo models. They can throw up things (often biased) observation misses. But to work these need good observed data. Baseball and cricket will better lend themselves to this-e.g. Steve Smith is more vulnerable to left-arm spin as the interaction is more bowler/pitcher to batsman rather than football/rugby where there are more interactions.
Statistics are important but football stats are just lists really. |
In order to draw any worthwhile conclusions form Monte Carlo models you need a lot of repetitions. There aren't enough, and you can't engineer enough, repetitions in football.
Supporters of the Moneyball approach will argue otherwise. | | | |
| This alternative table garbage on 19:45 - Sep 5 with 1454 views | Dewi1jack |
| This alternative table garbage on 21:39 - Sep 3 by jasper_T |
Typical responses from the anti-stats brigade.
Anyone who's watched our games knows we could easily have lost many of them. Our position in that table relative to the real one simply reflects that element of fortune and quality of finishing in our results so far. It's quantifying something that's should be obvious and observable to all of us.
Other teams are producing more chances and conceding fewer than us. That's an area for concern for Cooper going forward. Hopefully we can improve, create more and better opportunities, and sharpen up defending from open play back to last season's standards (when we under-performed relative to xG due to a terrible shot-stopping goalkeeper and lack of reliable finishers other than Oli).
But we're only 6 games in. Like the actual table you can't read too much into it yet. Wait until 10-12 games have been played. |
I know.
We're absolutely sh1te on the computer programs.
Shocking on fifa football manager or whatever game.
We should be relegated on stats and save our money.
Pity we're playing on grass and Fukin smashing it at the moment. 😂
And aren't I enjoying every moment.
Oh and I use SPSS on a daily basis before someone states I don't understand stats |  |
| If you wake up breathing, thats a good start to your day and you'll make many thousands of people envious. |
| |
| This alternative table garbage on 20:19 - Sep 5 with 1431 views | Tummer_from_Texas |
| This alternative table garbage on 18:58 - Sep 5 by perchrockjack |
Tummer.
Serious question,as they say.
When you see your country being ritually abused and insulted,how do you avoid seeing your backside .
We slag off America but,fook me,Trump or not, we love going there.
This is rank hypocrisy and we stink of it.
People from Uk will visit NYC,LA, even Houston without hesitation but baulk at a self drive holiday in Eastern Cape . |
It's a little unfortunate, but I don't think it's quite as bad as you say, and I don't take it personally.
Besides, anytime someone insults the USA around here, someone else usually insults them right back. | |
| |
| This alternative table garbage on 21:43 - Sep 5 with 1408 views | Professor |
| This alternative table garbage on 19:03 - Sep 5 by ploppy |
In order to draw any worthwhile conclusions form Monte Carlo models you need a lot of repetitions. There aren't enough, and you can't engineer enough, repetitions in football.
Supporters of the Moneyball approach will argue otherwise. |
Absolutely-too complex with 22 interacting players. Cricket or baseball- yes as although there are variables over 10 matches a batsmen may face a thousand deliveries. In football you don’t get this repetition. Performance analysis and GPS analysis make more sense. | | | |
| This alternative table garbage on 21:49 - Sep 5 with 1404 views | Dr_Winston |
| This alternative table garbage on 21:43 - Sep 5 by Professor |
Absolutely-too complex with 22 interacting players. Cricket or baseball- yes as although there are variables over 10 matches a batsmen may face a thousand deliveries. In football you don’t get this repetition. Performance analysis and GPS analysis make more sense. |
The repetition is the key.
Sports like Cricket and baseball largely consist of the same act (pitcher/bowler to batter/batsman) again and again and can be distilled down to stats far more easily than a fluid, creative game like football.
Trying to put a player like Messi or even a Celina in thrall to stats is absurd. They will only ever play a minor role in football. |  |
| Pain or damage don't end the world. Or despair, or f*cking beatings. The world ends when you're dead. Until then, you got more punishment in store. Stand it like a man... and give some back. |
| |
| This alternative table garbage on 21:53 - Sep 5 with 1400 views | Professor |
| This alternative table garbage on 21:49 - Sep 5 by Dr_Winston |
The repetition is the key.
Sports like Cricket and baseball largely consist of the same act (pitcher/bowler to batter/batsman) again and again and can be distilled down to stats far more easily than a fluid, creative game like football.
Trying to put a player like Messi or even a Celina in thrall to stats is absurd. They will only ever play a minor role in football. |
And one of the reasons we love the game. The unpredictable, the mercurial. Sadly being hammered out of rugby. | | | |
| |
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
