Stats

[BigbadBillyMcKinley]

Out of interest; does anyone have a clue or a suggestion as to how T K and his stat man, get the numbers for the players? I mean, you could watch a player have a blinder and have 100% passing accuracy and give them 20 for passing, but in reality they're a 5/6?
Do they watch them for a few games? A season?
I know they manage it on Football Manager, but the clear and obvious difference, is that that's a game and this is real life. I know that he uses the same system over yonder, but it clearly doesn't work. Jaguars are bottom of their resepctive table, we're 2nd from bottom of our respective table.

Is it all practise for this new wrestling things they've started up? All seem to involve tables.....

[Mitch]

It's more in depth than that for sure. Part of the reason we didn't take a young lad on loan from Man United was that due to his lack of game time they couldn't assess him satisfactorily. I don't know what the min. number of games are or if there is even one fixed number, but they will have some kind of scale.

[bog]

I heard that the procedure is based on Baseball. 

092.gif

[gezkc]

I don't know for certain, but I would expect there to be companies who collect detailed stats on every player in every league and clubs then buy this information and use it however they wish. I very much doubt Tony Khan or our scouts personally generate the stats themselves! It would need to be standardised and consistent across every league or else it's just meaningless.

Whenever you watch a football match, there are hundreds of stats shown about every player. I'm sure the TV companies also purchase their stat-based information too.

[r.harding]

This website is very good for explaining football stats, written for ordinary fans but by People in the industry. (A few were at Brentford advising a few years ago)... https://statsbomb.com
well worth flicking through their articles.

[filham]

Do we ever compare stats for a period before and after a player has signed for us. We can all imagine that the stats for some of our players this season must be pretty bad whereas they must have been good for us to have signed them.

You would think that tis simple comparison would destroy any faith in stats.

[Sting of the North]

Do we ever compare stats for a period before and after a player has signed for us. We can all imagine that the stats for some of our players this season must be pretty bad whereas they must have been good for us to have signed them.

You would think that tis simple comparison would destroy any faith in stats.

What? I truly cannot follow the logic in this at all? So, if we scout a player and then sign him and he performs badly then our faith in scouting would be destroyed as well? So, after deciding not to use stats or scouting, what should we do?

[MJG]

Do we ever compare stats for a period before and after a player has signed for us. We can all imagine that the stats for some of our players this season must be pretty bad whereas they must have been good for us to have signed them.

You would think that tis simple comparison would destroy any faith in stats.

What? I truly cannot follow the logic in this at all? So, if we scout a player and then sign him and he performs badly then our faith in scouting would be destroyed as well? So, after deciding not to use stats or scouting, what should we do?

If the scout says he's good and then we buy him and he fails to perform  we sack him as well.  How do we ever sign a player?

[toshes mate]

The data is not the fault and, as I have said many times before, data was used long before the advent of computers. 

The pathway to modern computer usage began around four decades ago and before that we had a variety of data collecting methods which would all end up in ink in one way or another e.g. typed, printed or written.   But what any crafts-person will tell you is that the brain is an almighty receptacle of experience gained both through trial and error, tutorials and mentoring.  In relevance to sport the craft-person was experienced in a discipline or maybe a few, knew what to watch out for, knew how to correct fault, knew how to improve performance, and knew how to evaluate outcomes. Although some of that could be passed on via the written word, a great deal more could not be passed on because it all requires experience of the real thing.

The computer database should be thought of in the same way as a really good book e.g. you won't enjoy it unless you thoroughly comprehend what is written, good, bad and indifferent.  The more books you read the greater will be your comprehension, and the more you understand the greater will be your demand for books written by decent writers.  That is the theory at least.  Data is just facts (e.g. gender) and figures (e.g. quantities) presented in raw form.  Only when you manipulate the presentation of the data by thinking through what you want it to achieve can you turn it into something that is useful, and many times you will have to accept the simple fact that what you have is absolutely useless.  The key is always understanding the data at the raw level i.e. how was it collected and how reliable is it for the purpose you intend to use it for.  There are genius kids and adults who can look at data and see patterns or sense in it almost instantly.  There are countless many people who will never understand what the data they are looking at means unless someone else refines it for them.

The data is always okay at some level and the computer will do what it is told, but the reader of the data and the operator of the computer may be forever flawed unless they have learned their craft via comprehensive experience of trial and error, tutorials, mentoring, and very great deal of practical work.  Unless they are genius of course.           

[Woolly Mammoth]

I heard that the procedure is based on Baseball. 

092.gif

Baseball is rounders, who plays that. Just blokes in costumes, no wonder our performances are a pantomime and a circus.
Too much emphasis on stats too much emphasis on dissecting every issue, and we still end up with injured homesick gutless snow flake players on huge salaries, who more often than not are in the Florence Nightingale Medical Ward at Motspur Park sitting on a bed pan.
So what do Fulham do, knife the bloke in the back who got us promoted in the first place with entertaining football, whilst the owners son, the real culprit who orchestrated and has been instrumental in undermining our club, gets away Scot free once again.
Did you know stats spelt backwards is stats.
Just like boy spelt backwards is yob. which originated from the phrase " backward boy ".
Exactly, it all falls into place, you guessed it, stats is a backward step, just like Fulham have taken, a backward step.
If you browse at the picture of me, you will note the words "Not One Step Back".
So they have ignored me, no wonder we are in a 2 and 8.
If Fulham are going to be backward enough that they have to use stats as a additional constructive help, then they must start using them properly. As the evidence so far is they are not.
It looks like it will cost us our membership in the Premier League and millions upon millions of English pounds in the process, that is not very bright is it.
So wake up Fulham stop sleep walking and blundering around like an enebriated backward boy.
Goodnight and thank you.

[Sting of the North]

The data is not the fault and, as I have said many times before, data was used long before the advent of computers. 

The pathway to modern computer usage began around four decades ago and before that we had a variety of data collecting methods which would all end up in ink in one way or another e.g. typed, printed or written.   But what any crafts-person will tell you is that the brain is an almighty receptacle of experience gained both through trial and error, tutorials and mentoring.  In relevance to sport the craft-person was experienced in a discipline or maybe a few, knew what to watch out for, knew how to correct fault, knew how to improve performance, and knew how to evaluate outcomes. Although some of that could be passed on via the written word, a great deal more could not be passed on because it all requires experience of the real thing.

The computer database should be thought of in the same way as a really good book e.g. you won't enjoy it unless you thoroughly comprehend what is written, good, bad and indifferent.  The more books you read the greater will be your comprehension, and the more you understand the greater will be your demand for books written by decent writers.  That is the theory at least.  Data is just facts (e.g. gender) and figures (e.g. quantities) presented in raw form.  Only when you manipulate the presentation of the data by thinking through what you want it to achieve can you turn it into something that is useful, and many times you will have to accept the simple fact that what you have is absolutely useless.  The key is always understanding the data at the raw level i.e. how was it collected and how reliable is it for the purpose you intend to use it for.  There are genius kids and adults who can look at data and see patterns or sense in it almost instantly.  There are countless many people who will never understand what the data they are looking at means unless someone else refines it for them.

The data is always okay at some level and the computer will do what it is told, but the reader of the data and the operator of the computer may be forever flawed unless they have learned their craft via comprehensive experience of trial and error, tutorials, mentoring, and very great deal of practical work.  Unless they are genius of course.         

Exactly. The fault is not with the data but with the interpreter of the data. The data in itself can of course be insufficient for the purposes for which you want to use it, but then again that is a problem with user not the data.

It is essentially the same as in scouting. The scouted player is playing exactly the same regardless of who is watching (i.e. the data is the same), but the interpretation of how that performance may potentially translate to a different setting will differ.

Both methods takes lots of experience, instincts and insight to ensure a good chance of success.

[Statto]

Both methods takes lots of experience, instincts and insight to ensure a good chance of success.

Agreed.

And that does beg a question for Tony Khan and some on here (not necessarily you) who seem to think it's acceptable for TK to have a key role in the stats process - would they be just as open-minded about him making himself head scout, going to games to watch players or our next opposition, etc?

There seems to be this false distinction between stats and scouting, whereby no one in their right mind would think someone with no football experience could suddenly be a useful scout, whereas someone can just dip their toe into the stats stuff without any knowledge or experience and they might be good at it, as if the only thing you need is some nerdy glasses (which naturally, TK has)

[S.F.Sorrow]

As far as I can tell from the Trumedia website (TK's stats company) they get their 'soccer' stats from Opta. Which is probably where most clubs get their stats (please correct me if I'm wrong)? If that's the case the difference would be the algorithms used to analyse the stats, and of course the people analyzing them. Whatever it is, something clearly isn't working right for us compared to other clubs.

It's also interesting to see that Trumedia have listed several high profile clients for American sports but as far as I can tell not a single reference to any client for 'soccer' (not even us). The fact that no high profile 'soccer' club seems to use them should tell us something. I have a feeling we are being used as guinea pigs for TK's company to break into the English/European market. Which has obviously been a massive failure with the results we've been getting.

Just my personal opinion of course, based on what I've found online.

[Woolly Mammoth]

I always felt that he/they had a hidden agenda, prior to purchasing of Fulham Football Club, and everything that has occurred since then has strengthened that analysis.

[The Rational Fan]

I don't know TK stats, but whoscored says these players have been our worst this season (with >180 mins playing time).
1. Stefan Johansen
2. Tim Ream
3. Kevin McDonald
4. Timothy Fosu-Mensah
5. Aboubakar Kamara
6. André-Frank Zambo Anguissa

I don't think the stats are so wrong. Some may want TFM and Anguissa as worse performer, but statistically they were ok before Everton, since then they have been the worse performers.

[The Rational Fan]

Do we ever compare stats for a period before and after a player has signed for us. We can all imagine that the stats for some of our players this season must be pretty bad whereas they must have been good for us to have signed them.

You would think that tis simple comparison would destroy any faith in stats.

What? I truly cannot follow the logic in this at all? So, if we scout a player and then sign him and he performs badly then our faith in scouting would be destroyed as well? So, after deciding not to use stats or scouting, what should we do?

If the scout says he's good and then we buy him and he fails to perform  we sack him as well.  How do we ever sign a player?

Statistcally, nearly every player in this squad is playing worse than they did last year, including the ones from the promotion winning team. The drop is far greater than could be expected with a player rising to the EPL from the Championshiop or Ligue 1. By example, most Cardiff and Wolves players statistics have only dropped by a whoscored rating of 0.25 to 0.35, most of our players have performance thats dropped by 0.4 (Odoi, MLM & Mitrovoic) to 0.8 (Ream, Johason, McDonald & Anguissa).

Recruitment is a problem, but FFC biggest problem is we don't even know why all the players aren't playing as well as they have in the past, age cannot explain it and saying their not up the standard is unknown (and in case of non-performing Schullre a little ridicilous as he looked more at home in the World Cup Final of 2014). Unlike Chelseas, we do not have near enough money as a club, for nearly every player to play worse than last season and even have a chance of staying up.

Next Season, we are likely to have almost the same squad (minus Sessegnon and Schullre), which would be one of the most valuable squads ever to grace the Championship and in my opinion should be an easiliy good squad enough to get promoted.

But, despite a decent Championship Squad, unless FFC improve the players we have i doubt this team would get promoted and FFC will end up the biggest underperformers in history and all our parchate payments gone.

Improving our recruitment is not going to help much as our next great recruitment drive will be if we get in the Premier League again. The only thing that will help FFC get back in the premier league is improving the players we have, but first we need to stop them getting worse.

[Mitch]

I don't know TK stats, but whoscored says these players have been our worst this season (with >180 mins playing time).
1. Stefan Johansen
2. Tim Ream
3. Kevin McDonald
4. Timothy Fosu-Mensah
5. Aboubakar Kamara
6. André-Frank Zambo Anguissa

I don't think the stats are so wrong. Some may want TFM and Anguissa as worse performer, but statistically they were ok before Everton, since then they have been the worse performers.

The stats don't lie. It's the interpretation that does.

Statistics don't sign a DM for £30m+ when we needed a centre back. More the problem for me than anything else. We tried to run before we could walk in this league.

[The Rational Fan]

I don't know TK stats, but whoscored says these players have been our worst this season (with >180 mins playing time).
1. Stefan Johansen
2. Tim Ream
3. Kevin McDonald
4. Timothy Fosu-Mensah
5. Aboubakar Kamara
6. André-Frank Zambo Anguissa

I don't think the stats are so wrong. Some may want TFM and Anguissa as worse performer, but statistically they were ok before Everton, since then they have been the worse performers.

The stats don't lie. It's the interpretation that does.

Statistics don't sign a DM for £30m+ when we needed a centre back. More the problem for me than anything else. We tried to run before we could walk in this league.

Jokavoic is on record on the 4th August 2018 saying we needed an ancrohman to cover McDonald. He didn't mention centre back, although we did buy Calum Chambers after this statement as a centre back.

https://www.standard.co.uk/sport/football/fulham-slavisa-jokanovic-calls-for-more-signings-transfer-business-definitely-not-done-yet-a3903926.html

[Mitch]

Fair point, but was it the most important rope? And is Zambo that player? Or £30m worth of that player? He's shown no aptitude to being a water carrier/defensive anchor whatsoever. He's better with ball at his feet keeping this moving than he is at breaking the ball up. We signed players we didn't need, spending big on those and then plugging the actual gaps with makeweights like Maxime LM and loans like Chambers and TFM.

Sent from my SM-G950F using Tapatalk

[Robbie]

Watch the film Moneyball. The obvious and massive flaw in our system is the assumption that European football is a single market.

As we well know an underutilized player in a mid table Bundes team is only like a half decent Championship player.