Christian Benteke

[Pirate55]

Are you playing the fool for trolling or do you really not get it? Serious question.

Serious answer. I do not doubt regression theory. I do not believe it can be applied meaningfully to footballers form to provide any worthwhile conclusions.

You on the other hand are so wrapped up in your world of statistics that common sense seems to not play any part in your equation. Bedford explained it for you in detail. There are too many variables. I have met many brilliant people that have no common sense, so don't feel too bad.

 

[milo]

I think you will find Scara thinks it can - see above.

I think that you are either intentionally misrepresenting what people have said or do not understand what is being said.

Of course regression to mean can be used as a tool to look at probability of future form. It cannot deal in certainties and no one is saying that it can.

 

[Rossi22]

The other thing with scara is that he's so transparently agenda driven. Like trying to compare AVB vs Sherwood and Soldado vs Ade using numbers heavily influenced by games against part time Russians or snow plowers from Lapland and pretending that it has any relevance whatsoever to Premier League form. It's not even comparing apples with oranges, more like apples with emperor king penguins.

 

[Pirate55]

I think that you are either intentionally misrepresenting what people have said or do not understand what is being said.

Of course regression to mean can be used as a tool to look at probability of future form. It cannot deal in certainties and no one is saying that it can.

If you really think it can, try using it to answer any of the questions Rossi has posed above.

 

[parklane1]

Sounds like this is turning into a baffle them with stats thread , at the end of the day he is a good player who fits for us and we should do whatever we can to sign him.

 

[Pirate55]

Sounds like this is turning into a baffle them with stats thread , at the end of the day he is a good player who fits for us and we should do whatever we can to sign him.

=D>=D>=D>

 

[spurs' magna carter]

Stats just state the obvious and is a FIFA/Football Manager way of approaching the transfer market.

Heat maps, tackles won etc bore me sense less, I used to read how Modric used to get little in the way of goals and asissts, something I went along with myself as I wasn't sure what he did. Until that is I saw him in the flesh (Villa H 10/11) and realised that whatever stats were produced he was the heartbeat of the team and by far the best CM we'd had since Hoddle and would most probably become one of the best in Europe, in his position.

If people don't want to sign Benteke then fine but if you cannot see the pure terror that he inflicts on defences when on the pitch, if you couldn't notice the difference on the pitch once he was sent off against us this season then it's probably because you were too busy looking at the stats.

There are somethings that stats can tell you that watching can't and vice versa.

Benteke would be a fantastic signing for us, whether he's sh!t at holding up the ball or not he scores goals a plenty and has proven himself in this league at a young age. The only reason why I wouldn't him is because of Kane, nothing else.

 

[parklane1]

Stats just state the obvious and is a FIFA/Football Manager way of approaching the transfer market.

Heat maps, tackles won etc bore me sense less, I used to read how Modric used to get little in the way of goals and asissts, something I went along with myself as I wasn't sure what he did. Until that is I saw him in the flesh (Villa H 10/11) and realised that whatever stats were produced he was the heartbeat of the team and by far the best CM we'd had since Hoddle and would most probably become one of the best in Europe, in his position.

If people don't want to sign Benteke then fine but if you cannot see the pure terror that he inflicts on defences when on the pitch, if you couldn't notice the difference on the pitch once he was sent off against us this season then it's probably because you were too busy looking at the stats.

There are somethings that stats can tell you that watching can't and vice versa.

Benteke would be a fantastic signing for us, whether he's sh!t at holding up the ball or not he scores goals a plenty and has proven himself in this league at a young age. The only reason why I wouldn't him is because of Kane, nothing else.

Great post k: and one I could not agree with more. As someone who has used stats at work they are a useful tool but they very rarely tell the full story, stats are fine but there are to many fans who think they are the only thing that matters.

 

[billyiddo]

^Stats may bore most of us senseless but those employed to make the big decisions in professional football will either assess them themselves or have a team of people to do so on their behalf. If you think anyone leaves things to gut instinct or intuition or whatever in this multi million(billion?) pound business you are niave to say the least

 

[scaramanga]

Serious answer. I do not doubt regression theory. I do not believe it can be applied meaningfully to footballers form to provide any worthwhile conclusions.

You on the other hand are so wrapped up in your world of statistics that common sense seems to not play any part in your equation. Bedford explained it for you in detail. There are too many variables. I have met many brilliant people that have no common sense, so don't feel too bad.

I understand all of that. If there's ever anything I need you to explain to me I'll let you know (likely to be somewhere between never and the end of the universe).

All of those factors are included in regression - they're what makes regression happen. I'll ask one last time because it really does become irritating for those of us that understand it - please stop posting about things you don't understand until you're willing/able to learn about them.

 

[scaramanga]

How do you know what the class of a 24yr old player is? Surely you can only make that judgement at the end of their career? What about Torres? What's his class? At what point did his run of terrible form, stop being form? What's the time frame? What's Kane's class? Is he going to regress soon? Maintain it? Get better? What does the big book of numbers say?

You know by watching them play - by seeing how they perform. If they have a sudden jump or dip they're unlikely to continue at that rate.

As much as we'd like Kane's rate of progression to continue I think only the most foolish of fans would expect him to continue improving at the rate he is. At his age he's likely to continue to improve but just more slowly than he has done this season.

 

[Finney Is Back]

My thoughts are that stats and probabilities absolutely can be used to determine player ability/effectiveness. The problem is that (so far) I don't think there are any models out there that can do it particularly effectively. If there were then there would be a club out there somewhere who never makes a bad signing. I think the problem currently is that hte models do not take in enough variables, or perhaps cannot yet properly correlate the variables. Big data is definitely the way forward for decision making in many industries but sampling and processing the big data and deciding what to do with it is still very much in it's infancy.

Personally I still think the best way of deciding whether or not a player is good/suitable is for the manager/scouts to actually go and watch that player.... and to do so on a number of occasions where they are playing in different circumstances and against different types of opposition. Otherwise it's all too easy to end up with players like Capoue who are a statistitions dream - making more passes than any other player in the league and also completing a record percentage of those passes. With the reality of the situation being that very few of them actually have any positive impact on the game with most of them simply allowing the opposition to get themselves organised back into their defensive positions.

 

[parklane1]

^Stats may bore most of us senseless but those employed to make the big decisions in professional football will either assess them themselves or have a team of people to do so on their behalf. If you think anyone leaves things to gut instinct or intuition or whatever in this multi million(billion?) pound business you are niave to say the least

Of course they do ( never said they did not) I have used them myself, they are a good tool but the point is they do not always tell the full story, those who uses stats as a definite are more then niave they are fooling themselves.

 

[Bedfordspurs]

How so?

Didn't achieve anything in their early parts of their career

Costas stats up to the beginning of last season were poor

Wright was playing non league

Lampard hadn't down anything until his second season at Chelsea

The reality is the statistical approach works if all this are equal... Football players don't work like that

 

[Bedfordspurs]

Out of curiosity... What would be the fixed variables for using in regression analysis in football

11 players each side at the start of the game (played get sent off and injured)

90 minutes if minimum playing time per team (extra time and injury time isn't constant)

Anything else?

Even the ball type is variable season to season

 

[Pirate55]

You know by watching them play - by seeing how they perform. If they have a sudden jump or dip they're unlikely to continue at that rate.

As much as we'd like Kane's rate of progression to continue I think only the most foolish of fans would expect him to continue improving at the rate he is. At his age he's likely to continue to improve but just more slowly than he has done this season.

Exactly this. Kane may or may not continue to progress. Bale did. Rebrov and Postiga didn't. Torres did and then didnt and now might. Ade is an anathema wrapped up in conundrum driven by contracts or juju. Soldado and Lamela - who knows at this stage? There is no probability one way or the other. Too many variables.

Benteke may or may not be a great signing. No one can know. And no one can predict. And there are no probabilities one way or the other. However, he has many attributes and has done it in this league. On a risk/reward basis he must ,on the balance of probabilities, offer us more chance of success than not. That is all.

 

[billyiddo]

Of course they do ( never said they did not) I have used them myself, they are a good tool but the point is they do not always tell the full story, those who uses stats as a definite are more then niave they are fooling themselves.

But is anyone using stats as a definite?

 

[Rossi22]

Scara is and does. I bet he's still bleating on about how it's pointless crossing the ball because only one in however many are converted .. as though that paints the entire picture. I'd love to know what stats are used to measure goalkeepers, defenders and defensive midfielders ... let me guess .. save %, tackle % and interceptions? :ross:

 

[scaramanga]

Exactly this. Kane may or may not continue to progress. Bale did. Rebrov and Postiga didn't. Torres did and then didnt and now might. Ade is an anathema wrapped up in conundrum driven by contracts or juju. Soldado and Lamela - who knows at this stage? There is no probability one way or the other. Too many variables.

Benteke may or may not be a great signing. No one can know. And no one can predict. And there are no probabilities one way or the other. However, he has many attributes and has done it in this league. On a risk/reward basis he must ,on the balance of probabilities, offer us more chance of success than not. That is all.

Benteke's had one good season in the Premier League - that's precisely why we need to rely on more than just looking at him and saying "he can probably do the job".

For your benefit, a list of players that had a stand out season who would have been a waste of money:

Benni McCarthy
Andrew Johnson
Kevin Phillips
Roque Santa Cruz
Michael Ricketts
Andy Carroll

That's why you don't spend £30M on a player who has had one good season - any old ****e can have one good season. If he has another like his first here then we can consider spending that kind of money.

 

[Bedfordspurs]

Would you take a player who has 1 in 2 in this league against a player who averages just over 1 in 4 in his career to date but not in this league (still top le el for most of it?)

Ones 24 and the other is 26.

One speaks English and the other doesn't ??