Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
xG
- Thread starter Groucho
- Start date
- Status
canuckblue
Player Valuation: £50m
Thought it was a shirt size (though for some on here that would be xxxxxG).
Martin Alvito
Player Valuation: £50m
It boils down to: if variance didn't exist, this is roughly the number of goals expected from the chances a side has created.
If you want a discourse on logistic regression, why that's the appropriate methodology, why maximum likelihood estimation is usually the method used to fit the curve and how switching up model parameters and methods of fit will cause discrepancies between various sources for xG, I can try to explain the basics...but you'll probably tell me to stuff it inside of five minutes, so I'll give that a pass.
The layman's explanation is pretty straightforward: the type of model being fit is much the same sort of thing medical professionals do when they're trying to figure out how much this risk factor or that risk factor affects things like mortality and morbidity rates, just without the time component they're dealing with. We're trying to predict a result with two states - goal or no goal, which is much like the medical professional trying to explain dead/alive or disease/disease-free.
The approach has problems. In principle, we should have horked up as many goals as Leicester (18-19) per xG. We have admittedly gotten lucky in a couple of games, but the model is positional in character and can't deal with things like good finishing, good goalkeeping and good blocks. Trying to do that results in trying to attribute more impact to players than we have enough information mathematically to justify, so the model has to sweep those attributes up in the variance term instead. You'll see more sophisticated models take defensive positioning into account, because that's doable, but player identity just isn't going to work because there's too many attacker/defender/goalkeeper combinations relative to the number of shots taken.
With a long enough data series, it starts becoming apparent that Messi was a very good finisher until he went to PSG. The model can't tell you why, though. It could be that the through balls he was getting were absolutely disgusting, it could be that he did dirty things to defenders off the dribble until recently, or it could be some combination of those or other factors. All it can say is that he was an outlier.
My advice would therefore to be to take any prediction of regression to the mean with a grain of salt, if you have cause to believe otherwise. xG will tell you that Haaland shouldn't be scoring as much as he is, but that's nonsense. He's just lethal, and he's proven that in previous seasons. It will tell you that we should have scored more goals than we did last season, but things like garbage long balls taken by a low-percentage shot-taker will fluff up xG without actual expected results. The model implicitly assumes that Gray is as likely to bomb in a worldie from distance as Cristiano Ronaldo, which we know isn't true.
TL;DR: xG is a good tool if you understand what it does well, and what it doesn't. Most people don't, and misuse it as a result.
If you want a discourse on logistic regression, why that's the appropriate methodology, why maximum likelihood estimation is usually the method used to fit the curve and how switching up model parameters and methods of fit will cause discrepancies between various sources for xG, I can try to explain the basics...but you'll probably tell me to stuff it inside of five minutes, so I'll give that a pass.
The layman's explanation is pretty straightforward: the type of model being fit is much the same sort of thing medical professionals do when they're trying to figure out how much this risk factor or that risk factor affects things like mortality and morbidity rates, just without the time component they're dealing with. We're trying to predict a result with two states - goal or no goal, which is much like the medical professional trying to explain dead/alive or disease/disease-free.
The approach has problems. In principle, we should have horked up as many goals as Leicester (18-19) per xG. We have admittedly gotten lucky in a couple of games, but the model is positional in character and can't deal with things like good finishing, good goalkeeping and good blocks. Trying to do that results in trying to attribute more impact to players than we have enough information mathematically to justify, so the model has to sweep those attributes up in the variance term instead. You'll see more sophisticated models take defensive positioning into account, because that's doable, but player identity just isn't going to work because there's too many attacker/defender/goalkeeper combinations relative to the number of shots taken.
With a long enough data series, it starts becoming apparent that Messi was a very good finisher until he went to PSG. The model can't tell you why, though. It could be that the through balls he was getting were absolutely disgusting, it could be that he did dirty things to defenders off the dribble until recently, or it could be some combination of those or other factors. All it can say is that he was an outlier.
My advice would therefore to be to take any prediction of regression to the mean with a grain of salt, if you have cause to believe otherwise. xG will tell you that Haaland shouldn't be scoring as much as he is, but that's nonsense. He's just lethal, and he's proven that in previous seasons. It will tell you that we should have scored more goals than we did last season, but things like garbage long balls taken by a low-percentage shot-taker will fluff up xG without actual expected results. The model implicitly assumes that Gray is as likely to bomb in a worldie from distance as Cristiano Ronaldo, which we know isn't true.
TL;DR: xG is a good tool if you understand what it does well, and what it doesn't. Most people don't, and misuse it as a result.
tommye
Player Valuation: £100m
Martin Alvito
Player Valuation: £50m
I would say that if the gambling degenerates don't care, they probably should. It will tell you things you know, like "if Wolves get somebody that can finish in, they're probably up" and things you don't, like "Bournemouth is bloody awful at chance creation - worse than Forest, and possibly much worse depending on who you ask."
It's not bullish on Leicester at all - some models have them as an even larger scoring outlier than City, and that's not likely to last since they don't have Haaland. It probably thinks Leeds is better than you do - better than Fulham, in fact. Of course, it also thinks there's not much separating Chelsea and Leeds either, and anyone with eyes knows Chelsea can play some defense...except at Leeds. That outlier is skewing things a bit.
It thinks that if Moyes and Emery sort those sides' finishing, they're top half. It thinks Brentford is legit. It thinks Brighton is some combination of unlucky and bad at finishing. The 9-0 shelling of Bournemouth deceives it into thinking the RS are better than they probably are.
Someone who was trying to make money off betting on footy, especially on the season-long bets, would be wise to learn how it works and why I'm drawing the above inferences. Used properly, it would probably help them spot odds-on opportunities where the betting public is likely out of sync with reality.
TheBigIguana
Player Valuation: £100m
No one has a bigger interest in the numbers than gamblers and then eventually the books. The principles behind xG started there I'm pretty sure, although the primitive version wasn't anything like where they are now.
One guy I follow posts a grid every week and it has the three promoted sides as clearly the 3 worst. It doesn't mean they'll all go down but if you're getting decent odds it is well worth taking.
Yea frank or some basic octopus... predicting games is hard, but if you're a gambler like I used to, or a pundit, you kinda put yourself in the position yo predict a world cup game. A lot of content is created solely around this futile exercise, its almost an industry in itself. So idk, might as well try to be good at it, and if you're going to try to be scientific about it, investing that time, I'd expect it to be better than what my dumba** off the street thinks.
Incel you can't even spell your name lol
So of course I'd expect this to go over your head like a belemic on a see-saw
Exactly, it's all about gambling. It has turned, not Xg specifically, but analytics into a useful tool for clubs, bur gamblers were the ones to originate and embrace this stuff I believe.
Why would you want a picture of a spliff?
This forums so [Poor language removed] weird
4 Eyes Better Than 2
Player Valuation: £8m
^^^ You reacted and liked this post from last week. Maybe stop being a fanny ?
This isn't the first time you've tried to pull Carlos for the things he posts.
XG assigns number value to the quality of an attacking position; basically it converts “chances” into a number. Sitters rate closer to 1.0. Slim chances rate closer to 0.0. Half chances are somewhere in the middle.
Imagine Chico makes an xV chart based on the likelihood that a multi has ever touched another person out of desire. If you get a thread with a few high xV multis, it’s expected to be a great thread.
Bulimic, right?
I knew I was right lol
- Status