I think what you can also learn from this is that football has too many variables for moneyball to be of any use and that xG is a pile of horseshit.
Announcement
Collapse
No announcement yet.
Thrown to three lions in a Roman arena - UKR v ENG
Collapse
X
-
Moneyball makes for an interesting strategy to build a Football Manager save around, and that's about it.
Originally posted by Rogin the Armchair fan View Post
Not half as much as it did for Brazil in 2016.
- Likes 1
Comment
-
Originally posted by Sean of the Shed View PostI think what you can also learn from this is that football has too many variables for moneyball to be of any use and that xG is a pile of horseshit.
xG has been shown to be a handy enough indicator in the case of a large sample size, such as all the games played by all the clubs over the course of a league season.
Obviously a finals tournament is a much smaller sample size.
Comment
-
Originally posted by scratchmonkey View PostI think England's low amount of attempts is an indicator of both how much they're looking to control the match and also how they've never had to chase a result. Also that the chances that they're getting are high-probability/falling to high-level players.
- Likes 1
Comment
-
Yes they aren't panicking and trying to force anything. Against Ukraine there was that one shot from Kane that was saved. I can't remember many other goal attempts. The last half an hour was about running the clock down, not giving the ball away and avoiding injuries and cards.
Comment
-
Originally posted by scratchmonkey View PostI think England's low amount of attempts is an indicator of both how much they're looking to control the match and also how they've never had to chase a result. Also that the chances that they're getting are high-probability/falling to high-level players.
In the past, Gerrard or Lampard would have been regularly shooting from 30 yards, hoping to score a wonder goal and England's fullbacks and centre halves would have been pumping meaningless highballs into the box for a half fit Owen or Rooney to fight two massive defenders for.
- Likes 1
Comment
-
Originally posted by Tratorello View Post
In the past, Gerrard or Lampard would have been regularly shooting from 30 yards, hoping to score a wonder goal and England's fullbacks and centre halves would have been pumping meaningless highballs into the box for a half fit Owen or Rooney to fight two massive defenders for.
Comment
-
"xG is shit" sounds a lot like the baseball die-hards over here who complain about how nerds have ruined a fine manly sport with their numbers and dorky eyeglasses.
I'm skeptical as to how useful it can be in terms of predictive forecasting for individuals; it can be pretty useful in terms of looking at a team and figuring out what's going on, two examples that spring to mind are Brighton of last year, where they should have scored a boat-load more goals than they did, now, that still leaves you in the situation of "are we not scoring enough goals because we're unlucky or are we not scoring enough goals because our players aren't good enough" (it's the latter) and it's still something that would be borne out just by watching them, you can see what's going on without going into calculations. The other example would be Liverpool and Manchester United the year where Mourinho got United to finish second -- United was the most overperforming team by xG, Liverpool the most underperforming. And regression to the mean followed, with United sliding down the table and Liverpool putting in what would have been a title-winning performance in almost every other year.
ETA: And of course, it has its most important application, which is in Football Manager, it makes it easier for me to tell my team that they really did alright and it's okay that they lost 3-nil because we had a higher xG, these kinds of games just happen sometimes.Last edited by scratchmonkey; 05-07-2021, 23:22.
- Likes 1
Comment
-
- Mar 2008
- 7495
- Off the purple line
- I'm slutty: Roma (on haitus until Jose is fired), Liverpool, and Dortmund
- Del Taco
I haven't paid much attention to analytics in football, but do pay a lot of attention to baseball analytics. There is a similar stat in baseball XWOBA (Expected Weighted On-base Average). For me, it's mildly useful in that people can tell if batters are having unlucky at-bats. They hit the ball hard but the ball keeps finding defenders. Baseball is a much more stat-friendly sport because good pitchers can locate the ball in quadrants where batters are likely to miss or hit the ball directly at defenders (that's not unlucky). Stats can tell us what batters do and what defenders do in comparison to other defenders. The sport is much more static. Football is much more difficult because players are always in motion and goals don't happen very often. We would only say that an attacker is unlucky if a keeper makes a truly exceptional save. Of course, players can improve, but I don't know if XG is going to explain anything that we don't see with our eyes. When Aubameyang was purchased by Dortmund, I thought Dortmund's scouts were crazy. His strike rate with Saint-Etienne was horrible. Then he came to Dortmund and was a world beater. I guess XG could have told us that he should have scored more but he was the one missing. Of course, the percentage of misses to goals scored when goals could have been scored would be a stat that would more neatly summarize his problems versus my anecdotal claim that his strike rate was poor. But whatever stat would be applied might ignore the midfielders and wingers who were feeding him or what he was expected to do in terms of pressing or dropping back.
Comment
Comment