- Status
catcherintherye
Player Valuation: £80m
It's a good post. I am a big fan of stats but also like just watching a player to get a feel for him. I probably prefer the latter in all honesty. It sounds a ridiculous thing to say, but often simply doing nothing can affect a game. The ability to stand still and hold your position. The ability to put yourself in a position that you are an option to be passed too and thus open up options. How do you measure that with stats?
Controversial though it is, but I've always thought both Carragher andNeville are astute judges of the game. Not on the basis of churning out facts, but understanding the intricacies of how the game is played. Carragher heaped praise on both Bernard and Gomes at Stamford Bridge for their ability to take the ball in difficult areas (again very hard to codify with numbers). He was ahead of the curve.
The amount of stats guys I see making out Gomes is just an ordinary Premier League midfielder (largely by looking at goal and assist stats). That he runs the game or Bernard can do similar is lost upon them and lost upon stats. Colin Harvey didn't score many goals (or get many assists as it happens) but he controlled the game and allowed others to prosper.
I remember Neville saying something similar in a debate about Paul Scholes, namely you don't just win a game by scoring a goal. You win game by controlling the flow of the game. It's very hard to measure something like that though. Yes you can use passes as a metric, but Scott Gemmill made a lot of passes, he never controlled a game though.
Sy-
Player Valuation: £8m
Yep, there's no stat for first touch. His high work-rate could possibly be measured by ground covered, but since it's up and down the line it would easily be lost in the data compared to a CM.
Both are integral to the flow of our moves and as a defensive unit respectively, furthermore there's the psychological affect of both on his marker(s) and how much deeper or over-committed they'd be because of this and his interplay, but now we're into completely unmeasurable territory.
As others have said, he should have a much higher assist number, but that's not on him.
D
Deleted member 28206
Guest
LOOKMAN DAVIES CITY GAME HELLO
Disgruntledgoat
Player Valuation: £50m
Nobody has yet come up with a good model for performance in a sport without repeatable patterns (pitching in baseball, downs in NFL even tackles in Rugby League) for a sport with non-specialized positions (IE there aren't rules governing what parts of the game they can influence at that moment). The work in this field in MLB and NFL is really advanced, but you don't see it in the NBA and those roles are much more defined than in football.
This is precisely because of the variables you identify here.
TheBigIguana
Player Valuation: £90m
At the same time if players like Gomes and Bernard are as vital as people say then the stats measuring overall team performance would look worse without them. Gomes for example missed three games after Fulham but as a team we played no worse. It's too small a sample to be meaningful but if something like that carried on it could show that the impact of the "unmeasureable" things is overstated.
Sy-
Player Valuation: £8m
Absolutely, it's about watching the player play and the subtleties of their contribution which is largely subjective.
Funny you mention Gomes vs Fulham, imo he was not up for any sort of battle that day which played no small part in that defeat. There are probably stats to suggest otherwise, and people probably blamed Gana for not picking up his slack, but watching the game itself he appeared to duck out of every single challenge (apart from the one he got a red for) as we got over-run and out-muscled.
zzr45
#MURDERINGSCUM
And for me it’s a clear point as to why Bernard was the better choice than Lookman - often you’d see Bernard peel off into a space and a defender would go with him because it was a threatening position - opening up more space for others. Lookman tended to drop back into pockets just off the wing, which while gave him lots of space, didn’t pull defenders out of position and didn’t particularly work as it would have often slowed the passage of play down.
There’s way and means of quantifying that, but as I was taught long ago, essentially when you start quantifying data it loses a lot of meaning - so even if it was classed as runs off the ball, given the example above they’d both be the same, but one would of been effective where the other had not.
Stats serve a purpose, they can be a decent starting point to analyse a team, but they are by no means an answer.
D
Deleted member 38674
Guest
In fairness, Gomes, Harvey and Scholes would never ever be defined by assists and goals because they aren't that type of player. They don't operate in the final third, so nobody would ever judge them based on that alone. There are still stats that accurately measure their performance though and for someone like Scholes in particular, you're able to see exactly why he's held in such high regard.
There are many on here who consider the use of statistics to be primitive, despite almost every successful business running on big data and every top club employing a team of stat analysts to collate and interpret this data. I appreciate that simply 'watching' a player in a given game (or even over 3 or 4) can garner more than a computer would, however this would even out over the season.
When taken across a long period, such as a season, statistics will paint a much more realistic picture than human memory ever could, especially when you take conscious / unconcious bias into account.
zzr45
#MURDERINGSCUM
The only measurable thing that matters is goals
zzr45
#MURDERINGSCUM
Here’s why what you’ve written is absolutely and totally incorrect. It’s from Das Reboot, a book exploring Germany’s win in 2014. In this passage a touchscreen system developed by SAP that relied on data analysis input is being discussed.
‘I worked with it, every single day,’ says Philipp Lahm. ‘That thing was amazing. It showed you everything you wanted to know about the other teams and your own team.’ As soon as you start playing around a bit on the touchscreen, it’s not hard to see why Lahm got hooked. You can navigate to games, teams or players. Important events–free-kicks, corners, goals–are synced with video footage. A couple of clicks, and you can see all of the chances Ghana created against the USA, all of Brazil’s free-kicks going back a couple of years, all key moments or any given position or any given player.
‘It’s not particularly clever,’ says McCormick, ‘but its ease of use made it an effective tool. We didn’t want to bombard coaches or players with numbers. We wanted them to be able to see, literally, whether the data supported their gut feelings and intuition.
It was designed to add value for a coach or athlete who isn’t that interested in analytics otherwise. Big data needed to be turned into KPIs that made sense to non-analysts.’
Clemens illustrated the point by sending back a slide with numbers from a World Cup knock-out game to Walldorf:
Opposition Goal Attempts: 18 On Target: 13 Germany Goal Attempts: 14 On Target: 12
A closely contested game, then. The 2-1 over Algeria? France? It was actually the semi-final against Brazil. The 7-1 win. ‘That just goes to show that a lot of football numbers don’t mean anything,’ says McCormick-Smith. ‘We concentrated on qualitative analysis instead. How do certain teams play? How do they defend corners, etc.’
TheBigIguana
Player Valuation: £90m
Ultimately yes. But the scoreline doesn't just happen. Thousands of things happen outside of the ball going into the net. And knowing which of those things teams that win more do and which teams that lose more do and figuring out how to play based on that is useful.
But football is football and often the results make little sense. That doesn't mean you can't try to understand it.
D
Deleted member 38674
Guest
It's very obnoxious to assert that someone else is indisputably wrong, and your 'evidence' to support that is an extract from a subjective book including the opinion of one Football player and cites a single anecdote about the incredibly poor interpretation of data in one particular game.
As I said, scouts will always be required because over a short period, you can notice things in a players game that don't often form part of the statistical gathering.
However, over the course of a season, if you want to get an idea on everything from a players fitness levels to their discipline, statistics will give you a better and more in-depth picture than any scout would.
zzr45
#MURDERINGSCUM
I like how you claim that someone else is obnoxious when you’ve been quick to lay claim to gobsh*te #1 crown round here.
Again, not just one book, not just one player. You’re just so incredibly wrong it’s actually funny. Carry on.
- Status
