A statistical review of Chris Weale

[akwahoo]

http://leicester.theoffside.com/football-101/season-preview-our-goalkeepers.html

Does the analysis made sense to any of you? I'm sorry if it makes no sense or I'm wasting your time. I am an American Foxile running the Leicester blog on The Offside (twitter: @allenjkha & @filbertway).

And Go Foxes tomorrow!

[Jordan]

I like where you're coming from, but IMO--though I'm no statistician--there are too many other factors involved in expected goals against for you to get enough information to isolate a goalkeeper's effect on the game. To be fair, though, 1) you did mention that and 2) to use one data point: he let in a howler against Coventry away, and that goal took him about one goal further than the total he would have expected to have conceded against them last season (that's a small sample size, but it's still noteworthy).

That said, Leicester could have expected to concede a lot of goals against Team A and were dominated in the games, but Team A could have blasted every one of their 50 total shots over the bar. Against Team B, though, Leicester would be expected to concede very few goals, but Weale could have been victimized by own goals. The lineups in front of Weale would also have a huge impact on goals against.

If you're going to go on the sabermetrics route, perhaps something like Range factor can be calculated for goalkeepers.

[akwahoo]

I like where you're coming from, but IMO--though I'm no statistician--there are too many other factors involved in expected goals against for you to get enough information to isolate a goalkeeper's effect on the game. To be fair, though, 1) you did mention that and 2) to use one data point: he let in a howler against Coventry away, and that goal took him about one goal further than the total he would have expected to have conceded against them last season (that's a small sample size, but it's still noteworthy).

That said, Leicester could have expected to concede a lot of goals against Team A and were dominated in the games, but Team A could have blasted every one of their 50 total shots over the bar. Against Team B, though, Leicester would be expected to concede very few goals, but Weale could have been victimized by own goals. The lineups in front of Weale would also have a huge impact on goals against.

If you're going to go on the sabermetrics route, perhaps something like Range factor can be calculated for goalkeepers.

Well said, and unfortunately there aren't any stats to show that (at least from what I've found). I'm not a statistician by trade either (although I understand sabermetrics), and I'm actually more towards the traditional side of player scouting and stuff like that. Just thought it would be interesting for a blog post.

Definitely agree on there being too many peripheral factors though. I really liked playing with the polygons though, that was fun haha. (Admit it, it looked cool to you too)

Looking forward to the match tomorrow (later today)!

[akwahoo]

I like where you're coming from, but IMO--though I'm no statistician--there are too many other factors involved in expected goals against for you to get enough information to isolate a goalkeeper's effect on the game. To be fair, though, 1) you did mention that and 2) to use one data point: he let in a howler against Coventry away, and that goal took him about one goal further than the total he would have expected to have conceded against them last season (that's a small sample size, but it's still noteworthy).

That said, Leicester could have expected to concede a lot of goals against Team A and were dominated in the games, but Team A could have blasted every one of their 50 total shots over the bar. Against Team B, though, Leicester would be expected to concede very few goals, but Weale could have been victimized by own goals. The lineups in front of Weale would also have a huge impact on goals against.

If you're going to go on the sabermetrics route, perhaps something like Range factor can be calculated for goalkeepers.

And ugh, I can't even try and fathom that howler. Le sigh...

[VampLCFC]

Is this for real?

[Sooper Steve's shin]

http://leicester.theoffside.com/football-101/season-preview-our-goalkeepers.html

Does the analysis made sense to any of you? I'm sorry if it makes no sense or I'm wasting your time. I am an American Foxile running the Leicester blog on The Offside (twitter: @allenjkha & @filbertway).

And Go Foxes tomorrow!

Great stats - Americans do this so well! We do try over here, but Sky Sports seem to think we can't cope with anything more than speed and distance of a shot. For the amount of airtime that football takes up in this country, I think there's a definite market for more detailed statistical analysis (the success of Football Manager shows that we all love a bit of stat action).

[Unabomber]

Took one look, got a headache so closed it.

[DJ Barry Hammond]

Interesting stuff, but a couple of things to consider...

1.) Have a read through again and you're spot some tweaks to make it read better and a couple of errors - by the way, who is Tom Flowers?

2.) You're stats boxes - I would have liked to be able to enlarge them off the page with a click. Maybe I could do it with more jiggery pokery, but it would be a nice touch to just click and bring them up bigger. Also the colour yellow for one of the lines is a poor choice on a white background, although I appreciate you've done it to represent Leicester colours.

3.) Your stats are based against on the main in respect of goals conceeded rather than Weales indiviudal contribution, so I don't know how much you can gauge is down to him and how much is down to those in front of him.

The clear difference between Football and Baseball is (although both are team sports) Baseball is much more about an individuals contribution to a team, where as football is more about the team working together towards the individuals performance.

In rudimentry terms a batter essentially has one thing to concentrate on and thats hitting the ball and relatively speaking none of his team mates can help him here which means statics in Baseball are much more revealing. However in terms of a goalkeeper, you can be the best in the world, but if you put him behind a pub team against Manchester Utd he is going to have a bad day in terms of goals conceeded?

4.) I appreciate this would have been a lot harder to come by, but I would be much more interested to see statisics of how goals were conceeded / shots saved percentage (and where these shots were going) / cross catch/punch/miss ratio and various other individual stats as that would have been more revealing to how good or bad Weale is.

As I said, this would have been a lot harder to come by and I expect if you were producing this kind of stuff you would be a professional statisican.

I wouldn't want to take away from what is a very interesting blog post (much better than, oooh look what I can than do on Footy Manager or the like) so keep up the good work!

[Jordan]

The clear difference between Football and Baseball is (although both are team sports) Baseball is much more about an individuals contribution to a team, where as football is more about the team working together towards the individuals performance.

In rudimentry terms a batter essentially has one thing to concentrate on and thats hitting the ball and relatively speaking none of his team mates can help him here which means statics in Baseball are much more revealing. However in terms of a goalkeeper, you can be the best in the world, but if you put him behind a pub team against Manchester Utd he is going to have a bad day in terms of goals conceeded?

Yes.

Football is moving more towards statistical analysis for a few reasons but perhaps one of them is to find a different way of evaluating players than watching clips of Iranian fullbacks on YouTube or turning or being subject to actually have to consider David Pleat's opinion just by turning on the television.

Baseball lends itself so much more to statistical analysis--hell, it's like the sport was made to be surrounded by numbers with decimal points. That said, though, baseball was stuck in the same thinking for generations (batting average, RBIs, ERA, wins and scouts saying "he's quick as lightnin', built like an ox and got an arm like a cannon"). "Sabermetrics" has revolutionized the way players are evaluated and compared.

Statistical analysis can have lots of merit. Moneyball: The Art of Winning an Unfair Game is definitely one of the best sports books one could read. It's about the late '90s/early 2000s Oakland Athletics--general manager Billy Beane was able to field very competitive teams despite having one of the lowest player budget and team payrolls in the league. Their secret was using sabermetrics to not only determine better ways to evaluate players, but more precisely, use those stats to figure out good players that were undervalued and cheap. If a football club--particularly a club that faces a major financial disadvantage from its rivals in the transfer market--could figure out something along the same vein, it could have a profound effect on football and the transfer market. Or, at least, when Milan Mandaric tells Manager X that he was going to buy Player Y, Manager X could point to some numbers that show Milan Mandaric is crazy about signing Player Y (and Manager X gets sacked).

It's happened before, err, sort of. Valeriy Lobanovskyi from Dynamo Kyiv was probably the most famous use-statistics-and-science-to-figure-out-crazy-shit-that-works manager. And then there was the massively successful Capello Index™

[fleckneymike]

Agree with the posts above. There isn't enough data out there to get a true picture of Weale's ability. Still an interesting thing to attempt.

You'd need to know both kicking and throwing accuracy and frequency (i.e. distance and location), shots to saves (again considering distance), catching against crosses faced (and again location of crosses), corners conceded, DFK and IDFK conceded etc.

As many on here would agree and disagree about there is a feeling that Weale struggles against shots from distance (and I'd add he gets beaten at his near post)

[Anish]

I can't say I'm a fan of detailed stats being used in football. The sport is a team game and there are so many variables that in depth analysis of statistics is in my opinion futile. Like Unabomber, I thought that blog looked more like a science experiment and I gave up halfway through. Stats can be manipulated and don't show a real picture - a midfielder could pass sideways and backwards all game long and have a 100% pass completion rate, but that doesn't make him the best midfielder if he can't pass forward or create anything for his team. Apparently Benitez used the Opta index stats to judge players and that's why he kept playing Dirk Kuyt - now in my opinion Kuyt is an above average player but no world beater, and just because he runs more than most and keeps the ball well (usually by passing sideways or backwards) it doesn't make him a quality winger or striker.

I don't mean to be offensive, but it's more an American trait to target statistics as the basis for judging teams or players. A top class goalkeeper could be playing for a pub team (or David James at Brizzle) but it doesn't matter how many shots he saves, he'll still concede shedloads and the stats wouldn't be an accurate picture when compared against an average keeper in a better team.

[akwahoo]

I can't say I'm a fan of detailed stats being used in football. The sport is a team game and there are so many variables that in depth analysis of statistics is in my opinion futile. Like Unabomber, I thought that blog looked more like a science experiment and I gave up halfway through. Stats can be manipulated and don't show a real picture - a midfielder could pass sideways and backwards all game long and have a 100% pass completion rate, but that doesn't make him the best midfielder if he can't pass forward or create anything for his team. Apparently Benitez used the Opta index stats to judge players and that's why he kept playing Dirk Kuyt - now in my opinion Kuyt is an above average player but no world beater, and just because he runs more than most and keeps the ball well (usually by passing sideways or backwards) it doesn't make him a quality winger or striker.

I don't mean to be offensive, but it's more an American trait to target statistics as the basis for judging teams or players. A top class goalkeeper could be playing for a pub team (or David James at Brizzle) but it doesn't matter how many shots he saves, he'll still concede shedloads and the stats wouldn't be an accurate picture when compared against an average keeper in a better team.

I mean I understand completely what y'all are saying, but I thought I'd give it a try since it looks interesting. But it's hard to take anything that's available and make any conclusions.

And P.S. I'm not a stats guy-- actually, I not the biggest fan of sabermetrics but understand them enough to see some importance. With the Capello Index and such gaining notoriety though, I thought I'd give it a shot. There's some interesting patterns. The Opta stats seem very interesting and insightful though.

--

When it comes to baseball, I'm completely about scouting and observing players more than numbers. I just understand the numbers' importance, so like I explained in the post, I think there is a blend. I'm just trying to think of interesting things, maybe you'll read my blog more haha!

And and my greatest sorry to the lad who got a headache, sorry about that.

[akwahoo]

It's happened before, err, sort of. Valeriy Lobanovskyi from Dynamo Kyiv was probably the most famous use-statistics-and-science-to-figure-out-crazy-shit-that-works manager. And then there was the massively successful Capello Index™

I really need to read the entire posts, you stole my thunder!

But thanks for being sympathetic, I was just trying to be interesting. And in fact, Foxesblogger has some more detailed stats that might be interesting for me. Maybe it'll reveal more, I will stay that I was partially motivated by Football Manager yes.

If anything though, I think the stats (which do give a picture of team defending at the moment) show that we need to be patient with our defense. I don't know if I believe in regression to the mean in football (unless it's just typical form going down), but our defensive form will surely get better with time.

[akwahoo]

Interesting stuff, but a couple of things to consider...

1.) Have a read through again and you're spot some tweaks to make it read better and a couple of errors - by the way, who is Tom Flowers?

2.) You're stats boxes - I would have liked to be able to enlarge them off the page with a click. Maybe I could do it with more jiggery pokery, but it would be a nice touch to just click and bring them up bigger. Also the colour yellow for one of the lines is a poor choice on a white background, although I appreciate you've done it to represent Leicester colours.

3.) Your stats are based against on the main in respect of goals conceeded rather than Weales indiviudal contribution, so I don't know how much you can gauge is down to him and how much is down to those in front of him.

The clear difference between Football and Baseball is (although both are team sports) Baseball is much more about an individuals contribution to a team, where as football is more about the team working together towards the individuals performance.

In rudimentry terms a batter essentially has one thing to concentrate on and thats hitting the ball and relatively speaking none of his team mates can help him here which means statics in Baseball are much more revealing. However in terms of a goalkeeper, you can be the best in the world, but if you put him behind a pub team against Manchester Utd he is going to have a bad day in terms of goals conceeded?

4.) I appreciate this would have been a lot harder to come by, but I would be much more interested to see statisics of how goals were conceeded / shots saved percentage (and where these shots were going) / cross catch/punch/miss ratio and various other individual stats as that would have been more revealing to how good or bad Weale is.

As I said, this would have been a lot harder to come by and I expect if you were producing this kind of stuff you would be a professional statisican.

I wouldn't want to take away from what is a very interesting blog post (much better than, oooh look what I can than do on Footy Manager or the like) so keep up the good work!

If you open the image in a new window/tab, it'll enlarge. I'd have to use some lightbox script to enlarge it here, and I don't know if my blog guy would let me do that now.

And yes, I absolutely agree with you on the deficiency of the stats. I understand the skepticism, and used the best of what I could find. I have some more advanced stats in my hand though now, so maybe the analysis will steadily improve. It's definitely tough and hard to look at stats and try and assign responsibility from them though.

[Cheese Me]

Love the idea and would definitely like to see more of the analyses on other players, agree with the point above that American's do this SO much better than us and that it makes interesting reading.

i don't think it really works for a keeper but I'm sure you'd be able to get much better conclusions for outfield players, especially midfield / forwards and contribution to goals scored which is more individual. Even for keepers though I think there's a lot more you could analyse more conclusively (crosses, claimed, shot stopping percentage, distribution to team mates, penalty saves, etc).

Thanks for the post though

[akwahoo]

Love the idea and would definitely like to see more of the analyses on other players, agree with the point above that American's do this SO much better than us and that it makes interesting reading.

i don't think it really works for a keeper but I'm sure you'd be able to get much better conclusions for outfield players, especially midfield / forwards and contribution to goals scored which is more individual. Even for keepers though I think there's a lot more you could analyse more conclusively (crosses, claimed, shot stopping percentage, distribution to team mates, penalty saves, etc).

Thanks for the post though

Any idea of where I can find some of these stats? I looked around but couldn't find these advanced goalie stats-- and I'm disappointed in myself for not being able to find them, they would be very helpful.

[Cheese Me]

Any idea of where I can find some of these stats? I looked around but couldn't find these advanced goalie stats-- and I'm disappointed in myself for not being able to find them, they would be very helpful.

Sorry but I haven't got a vlue. Maybe the Opta stats contain this?????

[Father Ted]

I find this sort of thing interesting. Perhaps more numbers could have been involved and percentages, comparing him to other goalkeepers in the league.

Although, without a solid statistical base I really don't see how you could do it for other outfield players.

Good luck though and I look forward to seeing how the next one comes along.

[Alexikokopops]

As a nerdy statistician I look forward to having a gander when I get a chance. As it stands I'm still in my pants and have 15 minutes to get ready for work.

[akwahoo]

As a nerdy statistician I look forward to having a gander when I get a chance. As it stands I'm still in my pants and have 15 minutes to get ready for work.

Hopefully it's alright, although I know that I made faulty conclusions attributing team success to judge a player. But I'm working on something bigger right now now that I have more stats available.