Carlos Carvalal form

[poite]

2 hours ago, ramone said:

No, Michael Ricketts and David Nugent have the same record so it would be unfair to say Jeffers is the best of all time when he has equals...

 

So you agree then. Using the stat that Jeffers has an 100% scoring record could indicate he's England's best ever goal scorer. But if you take other things into account I.e. that 2 other players also hold the same record, he only ever played one game etc it becomes misleading.

[ramone]

Just now, poite said:

 

So you agree then. Using the stat that Jeffers has an 100% scoring record could indicate he's England's best ever goal scorer. But if you take other things into account I.e. that 2 other players also hold the same record, he only ever played one game etc it becomes misleading.

Its a poor effort but if you like... 

[Mr. Tom]

2 hours ago, ramone said:

So by that very definition, we're not worried about the fact Atdhe Nuhiu hasn't scored a goal for Sheffield Wednesday since 23rd February 2016... because if you write: 0 goals in 569 days you can dress it up which ever way you like...? 

 

I get what you're saying mate, and i've probably substantiated your point with my example above but if you dont judge people on stats what do you judge them on? There are mitigating circumstances in everything you do in life so you have to draw the line somewhere and say fair do's. Well done Carlos. You have the best win percentage of any manager in SWFC's history. Yeah it might have rained a for a few more games under Big Ron which hindered his percentage bacause Steve Whitton never liked coming on in the wet. Or Marcus Tudgay didn't particularly like the new Mitre football they used in 2006 season so that had a negative effect on Brian Laws win ratio... Its swings and roundabouts. 

 

A yardstick has to be there adn the only way to do that is to keep it simple... 

 

 

Yep, you're right.

 

To give a bit more more context on what I posted re: stats - I think what I was trying to highlight is the way stats are so often used in football (not here, or by anyone in particular, just generally).

For some reason, football seems particularly vulnerable to the "correlation proves causation" fallacy, i.e. the idea that being able to spot patterns emerging between two independent variables (say, a manager and a win ratio, or a combined goals tally between two strikers) necessarily means they're influencing one another.

 

A commonly cited example of this statistical causation fallacy is the (genuine) fact that, in the UK, periods of higher ice cream sales nearly always correlate with increased rates of death by drowning. The way many people treat statistics - and especially in football - would be equivalent to saying 'therefore ice cream increases risk of drowning', which we can all clearly see is false. It's obviously just that more ice cream is sold in the summer, when more people go swimming. There's no direct causal relationship between the two, and yet the pattern is real, consistent and measurable.

 

I'm not in any way suggesting that's what's going on in this thread - again, I'm a big supporter of CC, I'm broadly very happy with how he's got us playing, and I do think he's drastically improved our results as a team.

 

I think I was just sort of playing devil's advocate in terms of the way stats are often used to imply a direct causal relationship between two things, when very often there can be all sorts of other factors at play.

 

Out of interest, (if you're not already bored rigid!), the full list of possible types of causal relationship between two variables, according to Wikipedia, is as follows:

 

- A causes B (direct causation);

- B causes A (reverse causation);

- A and B are consequences of a common cause, but do not cause each other (that would be our ice cream/drowning example, with the common cause being 'nice weather');

- A and B both causes C, which is (explicitly or implicitly) conditioned on. If A and B cause C, why do A and B have to be correlated?;

- A causes B and B causes A (bidirectional or cyclic causation);

- A causes C which causes B (indirect causation);

- There is no connection between A and B; the correlation is a coincidence.

 

...so in short, I'm basically saying that statistics are slippery little beggars, and can be used to prove almost anything if they're framed and presented a certain way. I've grown suspicious of them over the years, and now can't help myself - whenever I'm presented with a bunch of statistics, I'm always trying to think about what else could be going on!

Edited September 14, 2017 by Mr. Tom

[AwokenGiant]

21 minutes ago, poite said:

You've missed my point. The stats, as ramone pointed out are black and white and can't mislead. 

 

You've just proved my point. You can publish a black and white stat but without considering other things I.e. that he only ever played 1 game etc those stats can mislead

 

I understand your point and totally accept that there are a many number of different variables that can influence/skew stats. And, of course these variables should ALWAYS be taken into account when we are interpreting stats. 

 

However my point was to show that there has been an improvement on the field under Carvalhal and this is something that cannot be denied, for whatever reason.   

 

 

 

 

Edited September 14, 2017 by AwokenGiant

[ramone]

39 minutes ago, Mr. Tom said:

 

Yep, you're right.

 

To give a bit more more context on what I posted re: stats - I think what I was trying to highlight is the way stats are so often used in football (not here, or by anyone in particular, just generally).

For some reason, football seems particularly vulnerable to the "correlation proves causation" fallacy, i.e. the idea that being able to spot patterns emerging between two independent variables (say, a manager and a win ratio, or a combined goals tally between two strikers) necessarily means they're influencing one another.

 

A commonly cited example of this statistical causation fallacy is the (genuine) fact that, in the UK, periods of higher ice cream sales nearly always correlate with increased rates of death by drowning. The way many people treat statistics - and especially in football - would be equivalent to saying 'therefore ice cream increases risk of drowning', which we can all clearly see is false. It's obviously just that more ice cream is sold in the summer, when more people go swimming. There's no direct causal relationship between the two, and yet the pattern is real, consistent and measurable.

 

I'm not in any way suggesting that's what's going on in this thread - again, I'm a big supporter of CC, I'm broadly very happy with how he's got us playing, and I do think he's drastically improved our results as a team.

 

I think I was just sort of playing devil's advocate in terms of the way stats are often used to imply a direct causal relationship between two things, when very often there can be all sorts of other factors at play.

 

Out of interest, (if you're not already bored rigid!), the full list of possible types of causal relationship between two variables, according to Wikipedia, is as follows:

 

- A causes B (direct causation);

- B causes A (reverse causation);

- A and B are consequences of a common cause, but do not cause each other (that would be our ice cream/drowning example, with the common cause being 'nice weather');

- A and B both causes C, which is (explicitly or implicitly) conditioned on. If A and B cause C, why do A and B have to be correlated?;

- A causes B and B causes A (bidirectional or cyclic causation);

- A causes C which causes B (indirect causation);

- There is no connection between A and B; the correlation is a coincidence.

 

...so in short, I'm basically saying that statistics are slippery little beggars, and can be used to prove almost anything if they're framed and presented a certain way. I've grown suspicious of them over the years, and now can't help myself - whenever I'm presented with a bunch of statistics, I'm always trying to think about what else could be going on!

Wonderful post and to put your mind at ease, it kept me engaged all the way through. 

 

Your point is very valid in that stats can be used in which ever way them to be used. I've done it myself on many occasions. Maybe I should be a little more cautious myself...  

Edited September 14, 2017 by ramone

[Lincs Owl]

In my opinion, Gray turned us round and Carlos got us in top gear.  

The moaners maybe weren't going when we had Danny Wilson and the rest.