Statistics in football - Calling Archer, calling Archer

[david3103]

The Euro 2016 thread got a bit bogged down in the statistics about Goalkeeper's long-distance passing accuracy and it set me wondering about such stats and their usefulness.

For example, what defines an accurate long-distance pass?
Does it have to touch a member of our team? Or be collected and controlled?
If our keeper hits the same 5 sq yd piece of pitch that he aimed for, but our striker fails to make the right move, or just gets out jumped is that an accurate pass? Clearly it is, but is it counted as such?

Genuinely no tribal impetus to this, just curiosity.

Also, what stats can we consider as being genuinely meaningful? Pass success % for midfield players, chances created, goals scored are obvious but what stats are out there? Which sites repay the effort of getting used to viewing them.

[Archer]

I'll come back to this when I have more time.

For starters here is the table where  I picked-up the distribution stats filtered for goalkeepers  on whoscored.com which is one of several stats sites I play around with.

Click to see full-size image.

To be fair, stats sites are pretty useless for goalkeepers. For starters that one is in absolute terms and you have to work out the percentages manually to make it more meaningful.

whoscored take all the numbers from an Opta feed and the distinction between a short pass and a long pass is simply that the latter is for 25 yards or more. It is the same distinction for outfield players. I've no idea what the exact criteria is for determing success or failure. The only thing that table really tells me is that all goalkeepers struggle  with long ball distribution!

[Rod Paradise]

I've got to throw in a quote from Ebbe Skovdahl, ex-Aberdeen manager:

"Statistics are just like mini-skirts, they give you good ideas but hide the most important thing."

[david3103]

I've got to throw in a quote from Ebbe Skovdahl, ex-Aberdeen manager:

"Statistics are just like mini-skirts, they give you good ideas but hide the most important thing."

From The Bishop of Norwich:

"Statistics are like a bikini, what they reveals interesting, what they conceal is vital."

[Tal]

I've got to throw in a quote from Ebbe Skovdahl, ex-Aberdeen manager:

"Statistics are just like mini-skirts, they give you good ideas but hide the most important thing."

From The Bishop of Norwich:

"Statistics are like a bikini, what they reveals interesting, what they conceal is vital."

That would be Tim Westwood's dad, right?

[dino1980]

The Euro 2016 thread got a bit bogged down in the statistics about Goalkeeper's long-distance passing accuracy and it set me wondering about such stats and their usefulness.

For example, what defines an accurate long-distance pass?
Does it have to touch a member of our team? Or be collected and controlled?
If our keeper hits the same 5 sq yd piece of pitch that he aimed for, but our striker fails to make the right move, or just gets out jumped is that an accurate pass? Clearly it is, but is it counted as such?

Genuinely no tribal impetus to this, just curiosity.

Also, what stats can we consider as being genuinely meaningful? Pass success % for midfield players, chances created, goals scored are obvious but what stats are out there? Which sites repay the effort of getting used to viewing them.

Not to come across all 'Arbboy' but the biggest stat used to be wages. The correlation between league position and wages paid was usually pretty strong, way more important than the manager for instance.

But, assuming you mean actual in game football stuff then the two I think are most important are TSR% and Xg. TSR = Total Shots Ratio Ex: If Manchester City has 20 shots in the match and Newcastle have 10, City’s TSR for that match is .67, Newcastle’s is .33.

Xg = expected goals. Explanation here: https://differentgame.wordpress.com/2014/05/19/a-shooting-model-an-expglanation-and-application/

So, essentially outshooting opponent's and taking better quality shots are important by the looks of things.