Topic: WBA (H) Round Table

[csgreen]

You're looking in the wrong direction then.

Stop looking at the players as individuals, and look at them as part of a complex whole, where their part also affects the parts of those around them. That would be your starting point, rather than the almost totally fallacious "coin toss" analogy.

I'm really unlikely to be moved by the "almost totally fallacious" comment when there is honest to goodness data that suggests that performance is roughly independent across games (just like a coin toss).  Unless of course by "almost totally fallacious" you meant "supported by solid empirical evidence."  As for the complex whole - that's a perfectly reasonable hypothesis that is unsupported by any evidence that I'm aware of. Show me some evidence that an "on form" player improves the complex whole and I'll be glad to change my mind.  It's just that human are notorious for perceiving illusory correlations and the "hot hand fallacy" is amongst the most common...     

I'd could explain, but you're American and it would crumble your brain and then your brain would turn into cottage cheese. You then, would go on a mad firearms spree and give them all back and lobby your government to ban all weapons and then you'd lobby congress; to explain to your nation what a player 'being in form' means.
Then, then there would be a Tsunami of cottage cheese that would smash the shores of Ireland, who, unlucky for them, don't like Ryvita, so Ireland would then drown in cottage cheese. American cottage cheese!


Poor form by me there.

Given that I'm not sure any of that makes sense, I'm going to instead just mention that you improperly used a semi-colon.   

[PhaseOfPlay]

I'm really unlikely to be moved by the "almost totally fallacious" comment when there is honest to goodness data that suggests that performance is roughly independent across games (just like a coin toss).  Unless of course by "almost totally fallacious" you meant "supported by solid empirical evidence."  As for the complex whole - that's a perfectly reasonable hypothesis that is unsupported by any evidence that I'm aware of. Show me some evidence that an "on form" player improves the complex whole and I'll be glad to change my mind.  It's just that human are notorious for perceiving illusory correlations and the "hot hand fallacy" is amongst the most common...     

Care to point me in the direction of this data which shows that performance is roughly independent across games?

[DanA]

There's so many variables that make the coin analogy ridiculous. Confidence only a small part. Role in the team as POP pointed but even a clean run with injuries can make a huge deference. At a pro level just about everyone is carrying an injury of some sort during the season and that can have a huge impact on what a player can do. I often think a solid pre-season is the difference for a talented player.

[GrkStav]

Is perhaps the statistical analysis about 'clutch players' (so-called)?

That's not the same thing as players being in or out of form.

[ElstonGunn]

Care to point me in the direction of this data which shows that performance is roughly independent across games?

It's been pretty conclusively proven in both baseball and basketball that form has, at best, very very minor effects, and--in basketball at least--any "hot hand" is outweighed by players who think they have a hot hand taking bad shots and making themselves less efficient. Here's a review of the research in several sports.

I'm not aware of any research in football, but I will add that, before these studies, the most educated baseball and basketball experts were uniformly convinced that form did exist. And they still turned out to be wrong.

EDIT: I don't mean to say form completely doesn't exist (and the injury thing, I'm sure matters)--about half the studies find nothing, half find something very small. It's just not nearly as important as most think.

[ElstonGunn]

Is perhaps the statistical analysis about 'clutch players' (so-called)?

That's not the same thing as players being in or out of form.

No, there's been research on both phenomenons, at least in other sports.

[PhaseOfPlay]

It's been pretty conclusively proven in both baseball and basketball that form has, at best, very very minor effects, and--in basketball at least--any "hot hand" is outweighed by players who think they have a hot hand taking bad shots and making themselves less efficient. Here's a review of the research in several sports.

I'm not aware of any research in football, but I will add that, before these studies, the most educated baseball and basketball experts were uniformly convinced that form did exist. And they still turned out to be wrong.

We have different interpretations of form, then. "Form" for me, and accepted in the football circles I move in, is merely high-level consistency, peak performance and optional arousal state. What you are referring to and csgreen from the sound of things is the "hot hand" or streak? I would agree, that if that is the interpretation of form, then I could agree that it's almost random and independent across games. But if we're talking about a level were optimal performance is desired, consistently, and at the correct arousal state, then one would be wrong to argue that it doesn't exist, or that it isn't important. So perhaps the argument is more over a definition of terms, because what I consider "form" is certainly different to the idea of the "hot hand", which seems like a gambler's fallacy?

[ElstonGunn]

We have different interpretations of form, then. "Form" for me, and accepted in the football circles I move in, is merely high-level consistency, peak performance and optional arousal state. What you are referring to and csgreen from the sound of things is the "hot hand" or streak? I would agree, that if that is the interpretation of form, then I could agree that it's almost random and independent across games. But if we're talking about a level were optimal performance is desired, consistently, and at the correct arousal state, then one would be wrong to argue that it doesn't exist, or that it isn't important. So perhaps the argument is more over a definition of terms, because what I consider "form" is certainly different to the idea of the "hot hand", which seems like a gambler's fallacy?

It's definitely connected to the gambler's fallacy, but can you say more what you mean? You're ultimately talking about something that leads to improved performance, yes? Obviously the data won't be able to analyze the underlying physical and mental condition of the players, which I'm sure is very important. The basic idea is that the fact that Skrtel was good the last 3 games only gives you very slightly more valuable information than the fact he was very poor at the end of last year. I'll defer to csgreen, though, as he sounds much more immersed in this stuff than me.

[GrkStav]

No, there's been research on both phenomenons, at least in other sports.

I think the two phenomena, clutch players and 'streaks' or 'hot hands' are indeed closely-related.

I am cautious as to how the two relate to 'form' but also cautious not to 'dismiss' contrary evidence and arguments because I have a fixed idea (including from my own past playing experience) about the existence of 'form'.

I think we all have 'seen', or experienced in ourselves, 'bad form'. I suppose it could be normal, expected variation in performance, coordination, etc. that we conveniently label 'bad form'.

[DanA]

Why would anyone conduct a study on something which is so obviously true and why would we have such a superfluous study on hand?

A whole industry of betting for horse racing for instance is based around the form guide. I wonder would you bet against Sturridge scoring this weekend? Career average is about 1/3 and it's a tough away game, you should be massive favourite..... But for the form he's in.

[GrkStav]

It's definitely connected to the gambler's fallacy, but can you say more what you mean? You're ultimately talking about something that leads to improved performance, yes? Obviously the data won't be able to analyze the underlying physical and mental condition of the players, which I'm sure is very important. The basic idea is that the fact that Skrtel was good the last 3 games only gives you very slightly more valuable information than the fact he was very poor at the end of last year. I'll defer to csgreen, though, as he sounds much more immersed in this stuff than me.

My bottom line would be this: Do you think Suarez, for example, is in good form at the moment? Goal-scoring and otherwise? Most of us would impulsively answer in the affirmative. I don't know if that means that there's such a thing, unfortunately.

[GrkStav]

Why would anyone conduct a study on something which is so obviously true and why would we have such a superfluous study on hand?

A whole industry of betting for horse racing for instance is based around the form guide. I wonder would you bet against Sturridge scoring this weekend? Career average is about 1/3 and it's a tough away game, you should be massive favourite..... But for the form he's in.

Because it may not be true, even if it appears "obviously true". Such a study would definitely be "interesting" and thus likely to generate quite a good 'buzz' about its authors.

[PhaseOfPlay]

It's definitely connected to the gambler's fallacy, but can you say more what you mean? You're ultimately talking about something that leads to improved performance, yes? Obviously the data won't be able to analyze the underlying physical and mental condition of the players, which I'm sure is very important. The basic idea is that the fact that Skrtel was good the last 3 games only gives you very slightly more valuable information than the fact he was very poor at the end of last year. I'll defer to csgreen, though, as he sounds much more immersed in this stuff than me.

What I refer to as "form" is when players play up to the level expected for their position and unit within the team, and the relationships of trust that build up because of that consistency. So, for example, with Skrtel, there is no guarantee that he will play up to his current "form" in the next game - of course there isn't. But we can have a reasonable expectation that he will, based on the past few games. Additionally, he hasn't done anything to merit dropping from the team, because of the aforementioned "form" (or, in reality, consistency). Because of this consistency, his other two team-mates (Sakho and Toure) within the unit he is playing in, have come to understand his actions, the timing of his actions, and his decision-making, and they trust that he is consistent in these areas. So they have confidence in him, and that, in turn, breeds confidence in them from Skrtel (as well as his own self-confidence). So his arousal state is at an optimal level (not over-confident, not anxious, but calm and controlled), his performance remains high, his communication with his direct partners is at its best, and he shows "form" - i.e., he gives no reason to be dropped, and every possibility that this "form" or consistency will continue. There is no reason to expect his performance level to drop. What you lads seem to be referring to, is more concerned with forwards and scoring streaks, or keepers and clean sheets, where the outcome is clear (either/or) and one outcome doesn't influence the next, because the outcomes are largely random (as Reep and Benjamin, among others, showed in "Skill and Chance"). But this is different to the actual overall performance of a player, focused on processes rather than outcomes, because a large part of it is psychological, physical, and tactical in nature. It's why Carroll couldn't find "form", no matter how many games he played, and why someone as seemingly limited as Downing played a big part in our tactical play last season. "Consistency" is probably a better phrase than "Form", but the meanings are the same - a reasonable expectation, based on immediate past performances, of consistency in immediate future performances.

[ElstonGunn]

Why would anyone conduct a study on something which is so obviously true and why would we have such a superfluous study on hand?

A whole industry of betting for horse racing for instance is based around the form guide. I wonder would you bet against Sturridge scoring this weekend? Career average is about 1/3 and it's a tough away game, you should be massive favourite..... But for the form he's in.

Er, I linked to a review of 24 such studies in various sports above. Why would anyone do it? Probably because scientists like sports and think it's fun to study it and/or a journal is more likely to publish it because it's fun.

As for Sturridge, for his career he's more like 1 goal/180 minutes on the pitch, which includes a lot of time on the wing and as a teenager. His record when playing as a central striker is even better than that, and, none of what I'm saying implies that players don't improve. The last year or so of games Sturridge has played, where he's scored 20 in 26, are more relevant than his career numbers, since they strongly suggest he's either improved, is in a much better situation or both. So, no, I wouldn't bet against him scoring, but not because of his form.

[DanA]

Because it may not be true, even if it appears "obviously true". Such a study would definitely be "interesting" and thus likely to generate quite a good 'buzz' about its authors.

But it'd be proven in seconds. Look at Sturridge's career. He's a 1 goal in 3 games player but has scores 18 in 22 since at Liverpool. Surely that defies any reasonable normal distribution.

The strength and conditioning industry which has revolutionised sport is built around periodization. The idea that you can't constantly be at your physical peak and so must tailor training to peak at the right times. There's been thousands of studies on periodization and that alone disproves the idea that form is just random.

[GrkStav]

PoP, fair enough.

Now, given all that, how does a manager/coach deal with the following:

1. The 'covers' for the currently consistently performing starters getting 'rusty'
2. The 'covers' developing no 'synchronization'/chemistry/whatever-you-want-to-call-it with the the current consistently performing starters such that one or more of them can step and a. be in tune with the others b. the others are in tune with the new 'face'?

[ElstonGunn]

What I refer to as "form" is when players play up to the level expected for their position and unit within the team, and the relationships of trust that build up because of that consistency. So, for example, with Skrtel, there is no guarantee that he will play up to his current "form" in the next game - of course there isn't. But we can have a reasonable expectation that he will, based on the past few games. Additionally, he hasn't done anything to merit dropping from the team, because of the aforementioned "form" (or, in reality, consistency). Because of this consistency, his other two team-mates (Sakho and Toure) within the unit he is playing in, have come to understand his actions, the timing of his actions, and his decision-making, and they trust that he is consistent in these areas. So they have confidence in him, and that, in turn, breeds confidence in them from Skrtel (as well as his own self-confidence). So his arousal state is at an optimal level (not over-confident, not anxious, but calm and controlled), his performance remains high, his communication with his direct partners is at its best, and he shows "form" - i.e., he gives no reason to be dropped, and every possibility that this "form" or consistency will continue. There is no reason to expect his performance level to drop. What you lads seem to be referring to, is more concerned with forwards and scoring streaks, or keepers and clean sheets, where the outcome is clear (either/or) and one outcome doesn't influence the next, because the outcomes are largely random (as Reep and Benjamin, among others, showed in "Skill and Chance"). But this is different to the actual overall performance of a player, focused on processes rather than outcomes, because a large part of it is psychological, physical, and tactical in nature. It's why Carroll couldn't find "form", no matter how many games he played, and why someone as seemingly limited as Downing played a big part in our tactical play last season. "Consistency" is probably a better phrase than "Form", but the meanings are the same - a reasonable expectation, based on immediate past performances, of consistency in immediate future performances.

The argument isn't over whether there's a guarantee--we all certainly agree that there isn't one--it's over whether the bolded sentence is true. Like I said, I'm not aware of any research in football, and it would obviously be very difficult to measure, so I remain skeptical but agnostic. But I am talking about overall performance levels, not just either/or things like goals or clean sheets, though those are of course included. Baseball, as always, is the easiest sport to measure, and there it's pretty easy to see that form has an absolutely tiny effect on overall performance. It doesn't necessarily follow that the same is true in football, of course.

You do hint at something, though, which I think we all agree about. When you put a player in a better tactical situation, where the pieces compliment each other better, and the whole thing works, performance levels definitely go up. That's how I would explain Downing's usefulness. 

[PhaseOfPlay]

PoP, fair enough.

Now, given all that, how does a manager/coach deal with the following:

1. The 'covers' for the currently consistently performing starters getting 'rusty'
2. The 'covers' developing no 'synchronization'/chemistry/whatever-you-want-to-call-it with the the current consistently performing starters such that one or more of them can step and a. be in tune with the others b. the others are in tune with the new 'face'?

Small sided games where the units play as team-mates.

[GrkStav]

But it'd be proven in seconds. Look at Sturridge's career. He's a 1 goal in 3 games player but has scores 18 in 22 since at Liverpool. Surely that defies any reasonable normal distribution.

The strength and conditioning industry which has revolutionised sport is built around periodization. The idea that you can't constantly be at your physical peak and so must tailor training to peak at the right times. There's been thousands of studies on periodization and that alone disproves the idea that form is just random.

What you just said does not appear, to me, to have a valid connection to the existence or non-existence of "form". I don't see how periodization has anything to do with form, per se.

I could be misunderstanding your point, mind.

[ElstonGunn]

But it'd be proven in seconds. Look at Sturridge's career. He's a 1 goal in 3 games player but has scores 18 in 22 since at Liverpool. Surely that defies any reasonable normal distribution.

The strength and conditioning industry which has revolutionised sport is built around periodization. The idea that you can't constantly be at your physical peak and so must tailor training to peak at the right times. There's been thousands of studies on periodization and that alone disproves the idea that form is just random.

I already talked about the Sturridge example, but you're right with the second one. Fitness, etc. obviously does matter, so I think players certainly perform better and worse at different points.

So let me be more precise: I think the argument isn't that form doesn't exist, it's that you can't measure or predict it with anywhere near the precision that most think you can, and you certainly can't rely on immediate past performance (e.g. Sturridge's goal record) to identify form.

You also might say fitness and health are different from "form" (I think most would say this, though I'm not sure). The first undoubtedly is important, the second (i.e. being "in the zone"), it's unclear.

[PhaseOfPlay]

I already talked about the Sturridge example, but you're right with the second one. Fitness, etc. obviously does matter, so I think players certainly perform better and worse at different points.

So let me be more precise: I think the argument isn't that form doesn't exist, it's that you can't measure or predict it with anywhere near the precision that most think you can, and you certainly can't rely on immediate past performance (e.g. Sturridge's goal record) to identify form.

Let's go a little more philosophical then:

Do you think that the sun will rise tomorrow?

[GrkStav]

Small sided games where the units play as team-mates.

Bingo! That makes sense!

Do you know if Benitez uses those a lot? I vaguely remember something to the effect that he doesn't so much. Do I have that right?

[ElstonGunn]

Let's go a little more philosophical then:

Do you think that the sun will rise tomorrow?

That's a total non-sequiter. The question isn't whether the past is a good predictor of the future. Obviously Skrtel's past performances are indicative of how he will play in the future. The question is how much more we should weight his performances in the immediately preceding games than performances from somewhat longer back.

[PhaseOfPlay]

That's a total non-sequiter. The question isn't whether the past is a good predictor of the future. Obviously Skrtel's past performances are indicative of how he will play in the future. The question is how much more we should weight his performances in the immediately preceding games than performances from somewhat longer back.

That's not what I was reading from csgreen's initial post at all.

His initial post was that we shouldn't be picking Skrtel on "form", because form doesn't exist, and by implication, there is no reason not to pick Agger instead.

Did I miss something along the way?

[PhaseOfPlay]

Bingo! That makes sense!

Do you know if Benitez uses those a lot? I vaguely remember something to the effect that he doesn't so much. Do I have that right?

The sessions I've seen has some small sided games as part of them. But mostly unit training in functional drills, so the same principles apply - players in the same units working together with many reps.

[GrkStav]

Under what circumstances would you pick Agger over Skrtel, Pop? What must have happened or which opponents must be the next ones in order for Agger to be picked over Skrtel, given the current formation?

[PhaseOfPlay]

Under what circumstances would you pick Agger over Skrtel, Pop? What must have happened or which opponents must be the next ones in order for Agger to be picked over Skrtel, given the current formation?

For any of the three defenders - a basic mistake that costs a goal, at worst, or else just a below par performance where nothing is going right for the player.

Having said that, I'm actually in the horses for courses camp, with the exception of goalscoring and keepers. But I understand where Rodgers is coming from and there's a man-management dimension he has to always take into account since he has stated his preference for the player "in-form" to be selected, rather than players coming in on reputation

[Prof]

Ah... the old debate about the existence, or not, of form.

[CHOPPER]

Ah... the old debate about the existence, or not, of form.

Form is temporary, class is......



Well somebody had to say it!

[ElstonGunn]

That's not what I was reading from csgreen's initial post at all.

His initial post was that we shouldn't be picking Skrtel on "form", because form doesn't exist, and by implication, there is no reason not to pick Agger instead.

Did I miss something along the way?

The bolded is what I understood too. But that doesn't mean you don't rely on past information--in fact, you're relying on the past information of Agger's many years of IMO outperforming Skrtel to conclude he's better, and weighing that more heavily that Skrtel's good performance in the last couple of games.

Now, more recent information is always more valuable because it's undisputed that "true talent level" changes as players age, develop, etc., and that they'll be more or less effective in different tactical setups. So, for instance, Sturridge's performance since he joined LFC is (IMO) much more instructive than his total career numbers, which include a bunch of caveats, like being played out position and being very young. The question is how much more valuable is it. I'd argue that both Agger's and Skrtel's larger bodies of work are more predictive than Skrtel's recent good performances.

That said, I think it's mostly a man-management thing for BR, and I'm fine with that (even if it is a bit annoying to see Agger on the bench). The players I'm sure think form exists, so, in terms of man management, it doesn't really matter whether or not it's true.

[PhaseOfPlay]

The bolded is what I understood too. But that doesn't mean you don't rely on past information--in fact, you're relying on the past information of Agger's many years of IMO outperforming Skrtel to conclude he's better, and weighing that more heavily that Skrtel's good performance in the last couple of games.

Good points, but what a player did last year is not part of any definition of "form". Form is the recent past, whereas looking at the entire body of work is surely closer to "ability" or "quality" than "form"? In which case, the question "Is Agger better than Skrtel?" is the more pertinent question. But that's a different question to "Has Skrtel performed to a high level over the last 3-5 games?", which would be more indicative of "form"

Now, more recent information is always more valuable because it's undisputed that "true talent level" changes as players age, develop, etc., and that they'll be more or less effective in different tactical setups. So, for instance, Sturridge's performance since he joined LFC is (IMO) much more instructive than his total career numbers, which include a bunch of caveats, like being played out position and being very young. The question is how much more valuable is it. I'd argue that both Agger's and Skrtel's larger bodies of work are more predictive than Skrtel's recent good performances.

I would argue that their larger bodies of work are evidence of talent and ability and the corresponding gap between the two players - but that is still not the same as "form", which is more immediate past tense, and lends itself to a reasonable expectation of similar performance in the next game. Otherwise, why not pick Alan Hansen, for example? Or Robbie Fowler? But we know none of those would happen, because they haven't been top players in a long time. So "form" is both relevant and real - it's just a matter of definition, and I think csgreen is arguing from a pure statistical probability stand-point, which is valid, but fails to take into account psychology, systems, physical ability and tactical performance.

That said, I think it's mostly a man-management thing for BR, and I'm fine with that (even if it is a bit annoying to see Agger on the bench). The players I'm sure think form exists, so, in terms of man management, it doesn't really matter whether or not it's true.

Man-management is a huge aspect of it. In that sense, "reward for good performance" is probably a better approach to the issue.

[ElstonGunn]

Good points, but what a player did last year is not part of any definition of "form". Form is the recent past, whereas looking at the entire body of work is surely closer to "ability" or "quality" than "form"? In which case, the question "Is Agger better than Skrtel?" is the more pertinent question. But that's a different question to "Has Skrtel performed to a high level over the last 3-5 games?", which would be more indicative of "form"

I agree.
 

I would argue that their larger bodies of work are evidence of talent and ability and the corresponding gap between the two players - but that is still not the same as "form", which is more immediate past tense, and lends itself to a reasonable expectation of similar performance in the next game. Otherwise, why not pick Alan Hansen, for example? Or Robbie Fowler? But we know none of those would happen, because they haven't been top players in a long time. So "form" is both relevant and real - it's just a matter of definition, and I think csgreen is arguing from a pure statistical probability stand-point, which is valid, but fails to take into account psychology, systems, physical ability and tactical performance.

Again, this proposition, which you repeat as true, is what we're arguing about. Robbie Fowler has nothing to do with it--we have a lot of very good information that makes us think that his true talent level is now well below that of a PL quality player. csgreen is indeed talking about statistical probability, but I think you're making too sharp of a distinction between probability and the more subjective thing you say are factors. IF (and it's definitely still an if) immediate past performance is not predictive statistically of immediate future performance, then that means a good performance in the last game doesn't make you more likely to perform well in the next game.

Man-management is a huge aspect of it. In that sense, "reward for good performance" is probably a better approach to the issue.

Right, I agree about this.

[Prof]

It's ridiculous to debate the existence, or not, of form.

You could just as easily say that there is no research evidence proving the existence of God, or the F-lying Spaghetti Monster.  You can't prove the non-existence of something very easily, you can just say we haven't proved its existence yet.

The research around 'hot hands' is specifically looking at discrete skills, and the success rates of performing them.  The concept of form is much deeper than that and includes various components like psychological flow, perception, decision making, physical conditioning etc.

If we didn't talk about form, but about playing well, there is an abundance of research that shows that successful performance increases confidence, and confidence in turn increases performance.  It is also evidenced that poor performance can lead to loss of confidence etc.

[PhaseOfPlay]

 Again, this proposition, which you repeat as true, is what we're arguing about. Robbie Fowler has nothing to do with it--we have a lot of very good information that makes us think that his true talent level is now well below that of a PL quality player. csgreen is indeed talking about statistical probability, but I think you're making too sharp of a distinction between probability and the more subjective thing you say are factors. IF (and it's definitely still an if) immediate past performance is not predictive statistically of immediate future performance, then that means a good performance in the last game doesn't make you more likely to perform well in the next game.

So why pick teams then? Why buy expensive players? Why not just cobble together 25 lads for far less money, and hope that random chance turns them into champions? Because if a run of three good games does not lend itself to a REASONABLE EXPECTATION of a fourth good game, then the converse is also surely true - a run of 3 poor games doesn't mean the fourth will also be poor, so there's no need to drop a badly performing player, correct? Similarly, having a large body of quality work is also no predictor that Agger would do better than Skrtel if selected in his place, yes? So essentially, if form doesn't exist in any statistical instance, then managers can just throw names into a hat and select random combinations of 11 players for each game, with a 50% chance that they will be successful. But we both know that's nonsense. So why do some players get selected as ever-presents, and others don't? Why do some players score at a rate of 1 goal every 2 games, but a different striker in the same system scores only 2 in 20? Why do we select the higher scoring striker for each subsequent game, and not the low-scoring striker, if past performance is no indicator of immediate future performance? I know of the basketball studies that were performed in terms of the "hot hand" - but my challenge to that is - why would a player be selected with any regularity, if that is the case? If there is no statistical correlation between one performance and the next, then team-building is actually obsolete, it would seem.

[DanA]

What you just said does not appear, to me, to have a valid connection to the existence or non-existence of "form". I don't see how periodization has anything to do with form, per se.

I could be misunderstanding your point, mind.

Because I'm saying fitness among other things plays a part in form and if that is indeed the case. Then it instantly disproves the coin flip analogy  and the idea that form is perfectly random is proven false.


In the Skrtel/Agger argument I'd argue Agger's injury lost him his place and even before that he was making more mistakes than Skrtel. So why does he deserve to be played ahead of Skrtel when the defence right now is working. If it ain't broke don't fix it comes to mind.

[robgomm]

Why need there be a statistical representation of something for us to regard it as existing and relevant?

[JackWard33]

It's interesting Sturridge's record for us is being used in this discussion

When you look at his underlying shot statistics his goal scoring rate is totally unsustainable
He's actually a great example of 'the hot hand' (or running hot as gamblers put it) - or in football speak 'everything he hits is going in'

This doesn't mean he won't score a lot of goals for us it just means that his current 'form' isn't indicative of his future performance

[GrkStav]

Why need there be a statistical representation of something for us to regard it as existing and relevant?

There needn't. It's not a matter of statistical representation. And, in the end, you can only empirically disprove an empirically testable hypothesis.

Unless the 'form is important' argument can produce (logically derive) one or more empirically testable (and thus disprovable) hypotheses, then it remains 'per-scientific'. It doesn't mean it's false, at all.

[GrkStav]

It's interesting Sturridge's record for us is being used in this discussion

When you look at his underlying shot statistics his goal scoring rate is totally unsustainable
He's actually a great example of 'the hot hand' (or running hot as gamblers put it) - or in football speak 'everything he hits is going in'

This doesn't mean he won't score a lot of goals for us it just means that his current 'form' isn't indicative of his future performance

Suarez himself has spoken about this as affecting his goal-scoring rate. I do not think 'good' vs 'poor' goal-scoring streaks are psychological, unless one considers fine brain-motor coordination a psychological variable (which, I suppose it is, strictly speaking).

[robgomm]

There needn't. It's not a matter of statistical representation. And, in the end, you can only empirically disprove an empirically testable hypothesis.

Unless the 'form is important' argument can produce (logically derive) one or more empirically testable (and thus disprovable) hypotheses, then it remains 'per-scientific'. It doesn't mean it's false, at all.

I basically agree with this.

The hot hand argument would seem to me to be more relevant to how many headers or tackles Skrtl wins (ie: individual events) than his overall performance level.