No-one likes to see injuries but...

[crow87]

you seem to be saying a metric, with no supporting background information is valid. you don't even know what that metric really is, let alone saying what is right (or wrong) with it.

and that's without going into how there is zero evidence of a correlation...

Im going to call your bluff here, the above doesnt mean anything! And correlation between what exactly?
Anyway, the stats are clear. Its no use just using the "footy stats are rubbish" line, I'll be the first one to say they are (look at some of the good Dream Team players compared to how good they are as actual footballers) but these are clear cut. The stats are over a large enough sample of games, and the difference in output is crystal clear and by a significant margin.

 

[chopper55555]

I don't think Ken's limited in his mobility at all.

I hope so. I like Ken.

Just that Perrie is not good enough to play closer to goals and so to his credit takes the initiative and roams around to help the team in the best way he can. Kenny can also do it if needed.

Well... I think you'll find it's a game plan to have leading players down the ground... most good sides have them. Sarge is terrific at it if that's what he is. Kenny may do it however I think his value given his style may be more as the contested marker we've been missing under our lobby forward kick style.

 

[crow87]

Well... I think you'll find it's a game plan to have leading players down the ground... most good sides have them. Sarge is terrific at it if that's what he is. Kenny may do it however I think his value given his style may be more as the contested marker we've been missing under our lobby forward kick style.

Would'nt disagree with that for a second and its what I was getting at actually.

 

[Sanguinarius]

must say I'm surprised by the topic starter...

ken is a deep forward, sarge isnt his role is further up the ground.

so the question is who come in to fill sarge's role? if no one is there to fill that role of the roaming chf then our entire game plan changes. I'd back Stevens to fill that hole before kenny, simply due to his mobility.

 

[Crow-mo]

Im going to call your bluff here, the above doesnt mean anything! And correlation between what exactly?

you're going to need to know a lot more about this stuff before you call anyone's bluff.

as for correlation - well, your opening post made it very clear you don't know what that is. but in order to help you call my, and others, bluff in future: it is a relationship. a relationship between your data points that you think points to a conclusion.

Now if you had one of these, it wouldn't necessarily be correct - as a relationship is not the same as THE causal relationship or effect.

for example, 99.9% of all serial killers will have eaten white bread at one time or another in their childhoods - do we ban Tip Top?

however, you have not shown any relationship between either player and the outcomes you claim. or any outcome for that matter.

Anyway, the stats are clear. Its no use just using the "footy stats are rubbish" line, I'll be the first one to say they are (look at some of the good Dream Team players compared to how good they are as actual footballers) but these are clear cut.

lol. no champ, they are not clear cut. what they are is meaningless.

you have not controlled for opponent, weather, ground, midfield supply, forward line setup, F50 entries, overall team composition, positional changes and combinations, time on ground, coaching, injury, or any other of the 100's of things you would need to take account of before you could begin to make a case for anything.

in fact, they are barely stats at all.

And I love Kmac as a player and think the old Kenny should always be in our side if he's fit. He offers something different to Perrie, and when both players have been fit, up until recently they have both been in the side.

irrespective of the players used, this is not enough to even think about a relationship of the sort you imply.

The stats are over a large enough sample of games, and the difference in output is crystal clear and by a significant margin.

Both Perrie & Kenny had career years in 2005, playing together. that you do not seem to have even realised this, says all that is needed about your analysis and these "stats".

you didn't even bother to include any dates or timeframe, which is just shoddy.

what you did do, is clumsily try to pull some meaningless, unrelated data to further a bias against a player (again).

 

[NikkiNoo]

Spot on Crow-mo, there is a reason the saying "lies, damn lies and statistics" is still being used regularly

This discussion has made me realise that most of you actually don't know what our forward line structure is or how it works.

Here's a basic guide

3 talls and 3 mids/smalls (depending on available players).
Of the 3 talls at any stage during a game all of them will be the lead up forward up the ground. How often do you see Welsh outside the 50 taking a mark, wheeling around and sending the ball in? It is never JUST Perrie. Our forward line has a continual rotation on the field and on the bench.

If Kenny plays on Sunday I can guarantee that he will spend part of the game leading up the ground and not be based solely as a lead up forward in our 50. We have Welsh and Roo that are very good on a lead, why throw another player into there to clog up the leading space with them and their defender?

We try and maintain an open forward line with space for the forwards to use, this means that some of our forwards aren't in there, they are based on the 50 to then be able to run in and provide assistance and supposedly crumb the ball.

crow87 - if Welsh is 'lay-down Sally' why is he the leading goal kicker this year? I would have thought an effort was put in to mark and hunt the ball in order to kick goals? Agree that he has missed some gettable goals, but this is a guy that generally works his butt off to keep the ball in the area with his chasing of the defenders and putting them under pressure if not tackling them in order to turn the ball over. That statement you made makes me think that you really don't watch and understand our forward line, combined with your stats that as crow-mo pointed out in relation to 2005 when both Perrie and Kenny played in the team together and performed very well for both of them.

 

[Markthirtytwo]

Spot on Crow-mo, there is a reason the saying "lies, damn lies and statistics" is still being used regularly

This discussion has made me realise that most of you actually don't know what our forward line structure is or how it works.

Here's a basic guide

3 talls and 3 mids/smalls (depending on available players).
Of the 3 talls at any stage during a game all of them will be the lead up forward up the ground. How often do you see Welsh outside the 50 taking a mark, wheeling around and sending the ball in? It is never JUST Perrie. Our forward line has a continual rotation on the field and on the bench.

If Kenny plays on Sunday I can guarantee that he will spend part of the game leading up the ground and not be based solely as a lead up forward in our 50. We have Welsh and Roo that are very good on a lead, why throw another player into there to clog up the leading space with them and their defender?

We try and maintain an open forward line with space for the forwards to use, this means that some of our forwards aren't in there, they are based on the 50 to then be able to run in and provide assistance and supposedly crumb the ball.

crow87 - if Welsh is 'lay-down Sally' why is he the leading goal kicker this year? I would have thought an effort was put in to mark and hunt the ball in order to kick goals? Agree that he has missed some gettable goals, but this is a guy that generally works his butt off to keep the ball in the area with his chasing of the defenders and putting them under pressure if not tackling them in order to turn the ball over. That statement you made makes me think that you really don't watch and understand our forward line, combined with your stats that as crow-mo pointed out in relation to 2005 when both Perrie and Kenny played in the team together and performed very well for both of them.

I agree. It's a mix and match thing. I've even seen Pez in the half back line sometime. At that point I wonder what he's doing there until I have a closer look and notice he's not being an extra man in defence he is the link man to get the ball into our forward lines quicker. Then sometimes puts himself into a position to get the ball again and give it off to a running player going past.

A lot of people doen't see that.

Unfortuneatly last Saturday, that play unravelled us with his man staying home and not following him up the field and they then had an extra man in defence.

 

[crow87]

Firstly I must address and earlier criticism for not disclaiming where these figure come from, apologies I didnt mean to, for the record they're from the start of Craigy's coaching career at the crows in 2005 to present.

you're going to need to know a lot more about this stuff before you call anyone's bluff.

as for correlation - well, your opening post made it very clear you don't know what that is. but in order to help you call my, and others, bluff in future: it is a relationship. a relationship between your data points that you think points to a conclusion.

Now if you had one of these, it wouldn't necessarily be correct - as a relationship is not the same as THE causal relationship or effect.

Righto, what you're saying is a little hard to make sense of but anyway. Ok, this is stuff from around year 10 so I could forgive you for forgetting, (un)luckily they make us do statistics thumbsd:) as part of my course at uni so I've had a bit of a refresher. Anyway, back to basics:
In mathematical terms correlation defines how well two sets of data are related in our case this would be high values of one set of data correspond to high values of the other. In this case we're talking about averages rather than large sets of data, you could go back and work out the individual data points but i think you'd be wasting your time in this case. No footy stats are going to relate as clearly as something more mechanical but these do relate the respective players worths to the team rather well IMO. If you really want to bother taking the issue, we have high positive and negative correlation for both pieces of data in the original post.
If you mean correlation in rougher terms then I think its pretty clear too, for the stats that really matter (the score!) our scoring is 3 goals worse per game when Perrie plays than when he's injured, and 2 and a half times worse when Kenny is not in the team than when he plays.

however, you have not shown any relationship between either player and the outcomes you claim. or any outcome for that matter.

This doesnt make sense

lol. no champ, they are not clear cut. what they are is meaningless.

you have not controlled for opponent, weather, ground, midfield supply, forward line setup, F50 entries, overall team composition, positional changes and combinations, time on ground, coaching, injury, or any other of the 100's of things you would need to take account of before you could begin to make a case for anything.

Tell me why they're meaningless? They're over a large enough sample of games to mean that any of the above has a negligable effect on the data. Even if they do affect it 2.5 and 3 goal differences over this many games is substantial.

in fact, they are barely stats at all.

Right... say what?

irrespective of the players used, this is not enough to even think about a relationship of the sort you imply.

Uh-huh. I thought we'd established which two players we were talking about, why would the players influence the accuracy of the relationship? Im sorry but again its hard to see what you're trying to say. I think stats taken over more than 50 games for each player is getting close to enough for a relationship! At worst, a rough one, and even so look at the margin (repeating myself I know).

Both Perrie & Kenny had career years in 2005, playing together. that you do not seem to have even realised this, says all that is needed about your analysis and these "stats".

Most crows supporters would be well aware of this, apart from "SpringChoke" it seems. This isnt what Im talking about anyway, I see what Perrie does around the ground, how hard he works, links up etc etc and if Ken and Perrie were both in the team I would have no issues with it, what i do disagree with is that clearly Perrie has kept ken out of the team for a while now.

 

[crow87]

crow87 - if Welsh is 'lay-down Sally' why is he the leading goal kicker this year? I would have thought an effort was put in to mark and hunt the ball in order to kick goals? Agree that he has missed some gettable goals, but this is a guy that generally works his butt off to keep the ball in the area with his chasing of the defenders and putting them under pressure if not tackling them in order to turn the ball over. That statement you made makes me think that you really don't watch and understand our forward line, combined with your stats that as crow-mo pointed out in relation to 2005 when both Perrie and Kenny played in the team together and performed very well for both of them.

Obviously he is a very skilled player, it just frustrates me that he doesnt seem to always put in 100%, for a player of his skill level I think he wastes it a bit. Imagine if he put in the tackling and chasing efforts of a Porplyzia or a Shirley.

 

[NikkiNoo]

Obviously he is a very skilled player, it just frustrates me that he doesnt seem to always put in 100%, for a player of his skill level I think he wastes it a bit. Imagine if he put in the tackling and chasing efforts of a Porplyzia or a Shirley.

Maybe watch him a little more closely as that has always been an integral part of his game.

 

[marvin]

Maybe watch him a little more closely as that has always been an integral part of his game.

FWIW, I've watched him very closely the last two Saturday's, and it has not been an integral part of his game. His efforts have been sporadic in that area.

 

[King Elvis]

Welsh being our leading goal kicker probably says more about the rest of our teams than it does about Welsh.

As has been stated, my issue with Welsh is that he's so hot and cold. He can do nothing for an entire game, then have a decent last quarter and all is forgiven.

 

[Stiffy_18]

you're going to need to know a lot more about this stuff before you call anyone's bluff.

as for correlation - well, your opening post made it very clear you don't know what that is. but in order to help you call my, and others, bluff in future: it is a relationship. a relationship between your data points that you think points to a conclusion.

Now if you had one of these, it wouldn't necessarily be correct - as a relationship is not the same as THE causal relationship or effect.

for example, 99.9% of all serial killers will have eaten white bread at one time or another in their childhoods - do we ban Tip Top?

however, you have not shown any relationship between either player and the outcomes you claim. or any outcome for that matter.




lol. no champ, they are not clear cut. what they are is meaningless.

you have not controlled for opponent, weather, ground, midfield supply, forward line setup, F50 entries, overall team composition, positional changes and combinations, time on ground, coaching, injury, or any other of the 100's of things you would need to take account of before you could begin to make a case for anything.

in fact, they are barely stats at all.

And I love Kmac as a player and think the old Kenny should always be in our side if he's fit. He offers something different to Perrie, and when both players have been fit, up until recently they have both been in the side.

irrespective of the players used, this is not enough to even think about a relationship of the sort you imply.



Both Perrie & Kenny had career years in 2005, playing together. that you do not seem to have even realised this, says all that is needed about your analysis and these "stats".

you didn't even bother to include any dates or timeframe, which is just shoddy.

what you did do, is clumsily try to pull some meaningless, unrelated data to further a bias against a player (again).

I think you just ripped him a new one

 

[King Elvis]

Disagree stiffy.

The original statistics are still reasonably valid, they've both played basically during the same period of time.

Obviously what Crow-Mo is saying is correct, but he's demanding an incredibly strict definition of the collection of the stats, and for them to be thoroughly deconstructed to prove/disprove a point.

They aren't statistics to send a rocket to the moon, they're simple statistics which on the surface are reasonably relevant, and worthy of debate.

It's just a shame the thread has turned into a debate about the validity of the stats, and not the stats themselves.

 

[Crow-mo]

the original stats have NO meaning. zero. they apply a too simple approach to a more complicated issue, and in the process leave us with as little as you would expect.

you support them, because 1. you don't understand why they're (in)valid 2. you already have a pre-supposed bias, and have pre-supposed validity on anything that points to same, regardless of content.

simplifying something that is not simple might be convenient, but it's never going to have the slightest hint of accuracy.

 

[King Elvis]

the original stats have NO meaning. zero. they apply a too simple approach to a more complicated issue, and in the process leave us with as little as you would expect.

you support them, because 1. you don't understand why they're (in)valid 2. you already have a pre-supposed bias, and have pre-supposed validity on anything that points to same, regardless of content.

simplifying something that is not simple might be convenient, but it's never going to have the slightest hint of accuracy.

They aren't valid as a starting point for a discussion?

You're assuming I know nothing about stats - I'm not an expert, but I've done a bit of statistics; I didn't enjoy it and I have absolutely no doubt whatsoever that you've got more knowledge in the area than I do, but as much as you think I'm trying to simplify something, I think you're trying to complicate it.

Regardless, it's a moot point - Sunday afternoon we'll know whether Kenny is deserving of a spot in the side.

 

[crow87]

the original stats have NO meaning. zero. they apply a too simple approach to a more complicated issue, and in the process leave us with as little as you would expect.

you support them, because 1. you don't understand why they're (in)valid 2. you already have a pre-supposed bias, and have pre-supposed validity on anything that points to same, regardless of content.

simplifying something that is not simple might be convenient, but it's never going to have the slightest hint of accuracy.

Ok I think at the very least we've established that your posts have NO meaning. zero. You're nNot even willing to consider the stats as a starting point for discussion, whats the point of this forum then?

 

[crow87]

They aren't valid as a starting point for a discussion?

You're assuming I know nothing about stats - I'm not an expert, but I've done a bit of statistics; I didn't enjoy it and I have absolutely no doubt whatsoever that you've got more knowledge in the area than I do, but as much as you think I'm trying to simplify something, I think you're trying to complicate it.

Dont worry Elvis (never thought I'd say that in my life ), he seems to like to pretend to know alot while presume/instigate others know nothing. All the while his posts reveal an underlying ignorance on the topic (see stats 'discussion' above-not footy talk, thats obviously subjective).

 

[Crow-mo]

Ok I think at the very least we've established that your posts have NO meaning. zero. You're nNot even willing to consider the stats as a starting point for discussion, whats the point of this forum then?

lol. all we've established is that you're not doing well on a basic under-grad stats course.

oh, and that your performance there is still likely to be better than your understanding of our forward line

 

[crow87]

Disagree stiffy.

The original statistics are still reasonably valid, they've both played basically during the same period of time.

Obviously what Crow-Mo is saying is correct, but he's demanding an incredibly strict definition of the collection of the stats, and for them to be thoroughly deconstructed to prove/disprove a point.

They aren't statistics to send a rocket to the moon, they're simple statistics which on the surface are reasonably relevant, and worthy of debate.

It's just a shame the thread has turned into a debate about the validity of the stats, and not the stats themselves.

Thankyou, as I said in an earlier post, no footy statistics are going to be clear cut, its a sport for crying out loud, talk of "correlations" is simply laugable. Even so, they clearly point towards some sort of trend and deserve discussion. On an interesting and slightly ironic point, the stats course that I undertook, belittled by crowmo, is actually one required of people who want to send rockets to the moon (its called aerospace engineering), wouldnt have mentioned it but it tickled a funnybone in a strange sort of way.

lol. all we've established is that you're not doing well on a basic under-grad stats course.

oh, and that your performance there is still likely to be better than your understanding of our forward line

Oh please stop, I give in, thats simply too witty and destroying.

 

[Crow-mo]

They aren't valid as a starting point for a discussion?

not if they can't mean anything.

that is not to say that there are no stats relevant to footy, there are. but equally not everything applies in reverse.

for example, something you might be able to prove or at least suggest for discussion would be the output between 2 players. say, player X and player Y often play in the forward line together, in tandem.

you could put together their statistical output, and compare each when they play together, and when they are the on their own.
this could tell you whether they work together well, whether they don't, whether they possiblity get in each other's way, or whether there is any difference at all etc. etc.

only someone of very limited understanding however would try and attribute an outcome (total team score) that is not controlled significantly by the input (a single player) to that input.

it's flawed from the beginning.

You're assuming I know nothing about stats - I'm not an expert, but I've done a bit of statistics; I didn't enjoy it and I have absolutely no doubt whatsoever that you've got more knowledge in the area than I do, but as much as you think I'm trying to simplify something, I think you're trying to complicate it.

nothing complicated about something that doesn't fit.

Just as there is nothing complicated about when someone claims originally that they have meaningful statistics, but doesn't reveal anything about them, and hasn't bothered to do any analysis to establish a relationship.

that's the point when you know there is nothing there. oh, and when they throw a camp hissy fit. that too

Regardless, it's a moot point - Sunday afternoon we'll know whether Kenny is deserving of a spot in the side.

a fit kenny is always worthy of a spot in the side, you don't need statistics for that.

 

[crow87]

not if they can't mean anything.

that is not to say that there are no stats relevant to footy, there are. but equally not everything applies in reverse.

for example, something you might be able to prove or at least suggest for discussion would be the output between 2 players. say, player X and player Y often play in the forward line together, in tandem.

you could put together their statistical output, and compare each when they play together, and when they are the on their own.
this could tell you whether they work together well, whether they don't, whether they possiblity get in each other's way, or whether there is any difference at all etc. etc.

only someone of very limited understanding however would try and attribute an outcome (total team score) that is not controlled significantly by the input (a single player) to that input.

it's flawed from the beginning.




nothing complicated about something that doesn't fit.

Just as there is nothing complicated about when someone claims originally that they have meaningful statistics, but doesn't reveal anything about them, and hasn't bothered to do any analysis to establish a relationship.

that's the point when you know there is nothing there. oh, and when they throw a camp hissy fit. that too

Let me get this straight. You're saying that you can look at stats and determine all this, but statistics for the team scores from when player X and player Y play/dont play, taken from close periods of time, over a large sample of games cannot give any indication as to the players worth to the total that the team scores in a match.

a fit kenny is always worthy of a spot in the side

Ok now Im actually worried because I agree with something this guy said.

 

[Crow-mo]

On an interesting and slightly ironic point, the stats course that I undertook, belittled by crowmo, is actually one required of people who want to send rockets to the moon (its called aerospace engineering).

lol. ok. so why didn't you apply any of this "imaginary" learning?

I guess the imaginary teacher, didn't teach you that the statistical theory is the same, regardless of application. or didn't you do your imaginary homework that day.

Oh please stop, I give in, thats simply too witty and destroying.

obviously.

 

[Stiffy_18]

Disagree stiffy.

The original statistics are still reasonably valid, they've both played basically during the same period of time.

Obviously what Crow-Mo is saying is correct, but he's demanding an incredibly strict definition of the collection of the stats, and for them to be thoroughly deconstructed to prove/disprove a point.

They aren't statistics to send a rocket to the moon, they're simple statistics which on the surface are reasonably relevant, and worthy of debate.

It's just a shame the thread has turned into a debate about the validity of the stats, and not the stats themselves.

But the validity of the stats if exactly the reason why they are worthless to debate.

I mean seriously, anyone even with a slight knowledge of Introductory Statistics would know that these stats are just useless.

Crow-mo's comparions with serial killers and Tip Top is spot on.

 

[crow87]

lol. ok. so why didn't you apply any of this "imaginary" learning?

I guess the imaginary teacher, didn't teach you that the statistical theory is the same, regardless of application. or didn't you do your imaginary homework that day.




obviously.

Good stuff I got nothin'