Klaus wrote:As raw as a Catholic's mortal sin.
Statistics
Biggus wrote:Well raw data is just that- Raw data.
When you have to compile them it requires human input (GM's "sophisticated" statisticians) then all the personal prejudice misunderstanding and mistakes come into it.
Did the player fail to control a pass because he's useless or because it came at him with pace from close range? All the data says was he failed to control the pass......
Banduan asked a good question, whats the difference between a cross and a pass to someone in the center?
Whats the difference between a half chance a decent chance and a good chance?
That would depend on so many diverse things such as weather and the "type of game" it was........The statisticians have their opinions you have yours and I have mine, anyone can call on any statistic to prove any point they want, in football the only statistic that counts are the goals scored.
Pretty much anyone involved in football would disagree with you. That's why clubs invest so much in technology and sports science, as the stats gleaned from the variety of systems used do help.
You know as well as I do that football like any industry where money is thrown around, attracts the crooks drug pushers charlatans and snake oil salesmen.
Biggus wrote:Well raw data is just that- Raw data.
When you have to compile them it requires human input (GM's "sophisticated" statisticians) then all the personal prejudice misunderstanding and mistakes come into it..
No, you run tests on them. Human bias is taken out of it as far as possible.
Without statistics, we'd still be in the middle ages.
Oh yeah well then why don't we have to research it in Civ V?
You tell me.
Because statistics have also proved to be fucking useless when faced with rampaging Mongol hordes.
Biggus, presumably you've taken out insurance at some point in your life?
Probably not. Must seem useless in Biggus' lala land.
General Mirth wrote:Without statistics, we'd still be in the middle ages.
Thank Christ. A sane post in this thread.
With stats, you've got systematic measurements, and then interpretation of those measurements according to mathematical theory.
Without stats, you've got un-systematic measurements and then interpretation of those measurements according to whatever the fuck.
Statistics and probability are a huge powerup to empiricism and the reasoned study of cause and effect. Anyone who thinks stats are just a 'load of cobblers' needs their modernity licence revoked.
Note: this does not mean that I subscribe to every stupid interpretation of a set of statistics ever, they're just an essential toolāone that often provides a critical insight from an otherwise overlooked perspective.
Two main problems with the use of statistics in football.
Definition - basically what Biggus says and also what I have said. You can use them easily for basic arguments like "We don't get enough corners" or "[X] is always offside". A 'corner' and 'offside' have clear definitions. It's when they're used for arguments like "X doesn't create enough chances", "Y doesn't convert enough easy chances", "...% of our goals come from the right side" that we go into dodgy ground. Chance creation or shot conversion rates etc. you can't take seriously.
It's not actually the fact that you're dealing with human error that makes statistics in football difficult. It's the low frequency of games. Basically, for what you're trying to analyse or predict, your sample size is far too small. Man Utd play Arsenal (usually, these days) twice a year and with that you have team changes and all sorts in the interim. In financial markets (which also deals with human behaviour), stock prices change (almost) continuously. It's easier to look at the last value and base a quick prediction on that than it is in football.
I'm with Biggus in that these statisticians hired by clubs are regression monkeys that couldn't hack the hardcore stuff. A lot of their 'analysis' is probably garbage. Football doesn't easily lend itself to (good) statistical analysis.
I agree with that.
Statistics and sports in general don't offer the same clarity as they do in it other fields. And the numbers that we, as fans use, is nothing more than basic mathematics. I would think clubs actually do more than look at the trivial stats but like you said the numbers would only serve to eliminate obvious outliers.
My contention is Biggus' misunderstanding of what statistics is and his (frankly astonishing) claim that it is a niche discipline.
What???
You just admitted that football doesn't lend itself very well to statistical analysis (my point)
Then you reject my contention thats it's a niche discipline, which is it- make up your mind.
Statistics may be useful for finding out what 9/10 cats prefer but thats about it.
Biggus wrote:What???
You just admitted that football doesn't lend itself very well to statistical analysis (my point)
Then you reject my contention thats it's a niche discipline, which is it- make up your mind.Statistics may be useful for finding out what 9/10 cats prefer but thats about it.
I think your misunderstanding Biggus is you think statistics is limited to statements such as ā3 out of every 4 car accidents occur after 9pmā or āArsenal are just 2% away from dominationā.
The rest of us are thinking of statistics as a sub-discipline of mathematics. I'm a maths researcher (particularly in statistical theory) and I don't take kindly to being called a crook or a snake. I don't con anyone. Its methods and models are used in estimation problems everywhere.
Biggus wrote:What???
You just admitted that football doesn't lend itself very well to statistical analysis (my point)
Yes. But not for the reasons you outlined.
Then you reject my contention thats it's a niche discipline, which is it- make up your mind.
Just because it doesn't apply to football doesn't make it niche. Don't be difficult, there is no contradiction.
Your contention:
Biggus wrote:Statistics suck! they're just for geeks and anoraks who've never gotten their knees dirty in their lives.
When you have to compile them it requires human input (GM's "sophisticated" statisticians) then all the personal prejudice misunderstanding and mistakes come into it.
Both of which is wrong and, crucially, not how statistics works!
Statistics may be useful for finding out what 9/10 cats prefer but thats about it.
Statistics is used in everything from psychology, to medicine, to finance and engineering. There would be no research, no ability to understand large volumes of data without statistics! Hence the middle ages comment because people would be stuck believing superstitious nonsense without probability theory.
Statistical testing is what takes bias out of our subject matter. Not the other way around.
Look I had a similar discussion with Kel years ago on A-M you might recall, it concerned chance and mathematical probability.....
Statistical testing will tell you that if you have one hand in a bucket of boiling water and the other one in ice water you'll be perfectly comfortable.
Yes our world seems to be based on physical laws and its a comfort to some that life can be made more predictable and logical, but I've news for you- We've never left the middle ages- We've never even left the stone ages, we are still creatures who make their choices on emotion prejudice superstition and irrationality.
In a static controlled unchanging logical environment you can rely more on statistical and mathematical analysis but once humans become involved the doors to chaos are opened and there are too many variables that can't be predicted- The ghost in the machine.
Sounds great but that's empty bluster. Sorry, but there's no sense in taking this any further if you don't do a little homework first because you clearly don't understand the subject.
Biggus wrote:Look I had a similar discussion with Kel years ago on A-M you might recall, it concerned chance and mathematical probability.....
Statistical testing will tell you that if you have one hand in a bucket of boiling water and the other one in ice water you'll be perfectly comfortable.
I don't know what this means. Just give me one example of a well-used test statistic which is so flawed. No-one would take averages seriously in this extreme example. And more to the point - this hardly relates to the real world. Strikers don't score 40 goals in one game and 0 over the other 37. If a striker has a goal-a-game average, it's usually because they've scored regularly. And if their goals do come in patches, this will be noticeable.
Yes our world seems to be based on physical laws and its a comfort to some that life can be made more predictable and logical, but I've news for you- We've never left the middle ages- We've never even left the stone ages, we are still creatures who make their choices on emotion prejudice superstition and irrationality.
In a static controlled unchanging logical environment you can rely more on statistical and mathematical analysis but once humans become involved the doors to chaos are opened and there are too many variables that can't be predicted- The ghost in the machine.
This is often a criticism of economic analysis: 'too many unrealistic assumptions'. The reality is that - unless particular distributions and assumptions are considered completely appropriate (and they can be in large samples) - nonparametrics methods are used which set out to impose no conditions on people acting rationally and, well, minimal assumptions in general. Estimation and inference is completely based on what actually happens. Whether humans react irrationally or rationally, it doesn't matter. All that shows up is what they do. And when you have repeated observations over many individuals and/or time, the accuracy of estimation can be powerful.
General Mirth wrote:Sounds great but that's empty bluster. Sorry, but there's no sense in taking this any further if you don't do a little homework first because you clearly don't understand the subject.
And maybe I should wait until you've lived a bit longer and realised-
Data is not information
Information is not knowledge
Knowledge is not wisdom.