The only one I'm sure of is the last one because the defence was hilariously awful that season.
United and xG (now that Ole is gone will things change?)
- Thread starter Pogue Mahone
- Start date
-
In the latest episode of the United Hour Podcast, Nik and Imran hold a long overdue conversation about United's awful league form with 1 win in 7. They discuss a disappointing draw at Bournemouth, a not so disappointing draw against Liverpool and a crushing defeat at Chelsea - Is time nearly over for Erik Ten Hag?
Then in a special second half of the pod, Nik welcomes Natalie ( https://twitter.com/natalie_17) to discuss the fortunes of Manchester United Women as they booked a place in the FA Cup Final after a historic first ever win vs. Chelsea.
That's 3 out of 5.
Swap 2 and 4 I think. We were terrible in 2017/18 even though we finished 2nd. I remember that's when the xG arguments started on here as well
I had a look now and you're correct about this, 19 points over the xPts are insane though.
Yeah, that 2nd place was quite deceptive. Last season's 2nd place wasn't quite as extreme but still, our overall results clearly and significantly flattered us.
It's mad that in terms of overall performances (ignoring results), we still haven't surpassed 2016/17 - and that season was nothing special, really, and we finished effing sixth.
I think we were really good that season. A lot of games where keepers had an insane game vs us. Heaton one of them.
Plus we stopped giving a feck about the league after Swansea/Bournemouth draws
PSG won 2-0 today with an xG of 0.48. The guys in here should be telling their fans to start frothing at the result right about now.
Obviously not but they're unlikely to win the CL if they perform like that against the top sides in every game.
Exactly. 5 minutes since winning against one of the best teams in the tournament and already crying about what might happen in the rest of the games.
Who is crying?
PSG fans obviously. Their team just lost by 1.5 in xG.
If they do this every game and win the CL in this manner then gloating is more deserving, but what are the odds they can ride out their luck this way?
I don't get this - are you dismissing xG or praising it?
Are you actually asking this question about a club tournament that was won by Inter in 2010 and Chelsea in 2012?
You've only listed 2 sides from the last 12 years.
And I didn't say the odds are zero. They are smaller.
It’s possible just not probable.
You can help yourself to other club and national teams who have won CLs and WCs that went completely against xG.
Pretty much the point I was making. A 2-0 winning result is less important than the probability of getting the result, after the result. Mental.
Eh?
Feel free to celebrate the result if you're a PSG fan but any neutral observer notes that
1. They rode their luck in this win (this wasn't tactical excellence like Inter 2010), and
2. If they give this performance again against top opposition they are very likely to lose more than they win
And there's a plethora of actual past evidence that laughs in the face of a prediction like that.
I don’t think there is because as far as I’m aware there isn’t xG data for any of the examples that you cited. You’d really need to look at that for both Chelsea and Inter’s entire run to see how often they really got away with it.
Do you actually need xG data for those Chelsea Barca games? Christ on a bike.
That’s one or two games and yes you’d still need to look. They may be have been good value for the win in their other games. How many ‘lucky‘ sides do you think have won the CL compared to ‘deserving’ winners?
There's nothing lucky about winning games while having lower xG than the opponent. Plenty of teams literally plan entire seasons based on that strategy and have come out with league wins, let alone CLs or WCs. Chelsea won back to back leagues by parking the bus and playing 2 DMs with a flat back 4 and they'd win those leagues 100 times over if it were repeated. 'Smaller probability'.
Imagine if this lot was around after the CL semis in 07/08 when we defended for pretty much the entire second leg won 1-0 on aggregate from a 30 yarder rocket from Scholes. They'd have probably cried themselves to sleep the night we got to the final.
'More likely' doesn't mean that the 'less likely' thing never happens. It's just... less likely to.
In football, the team that plays better and creates more and better chances than the opponent wins more often. But not always.
(also, parking the bus doesn't necessarily mean you have lower xG than your opponent)
That's fine even if it means very little given we aren't talking about vast differences in probability here. Teams that create far lesser chances than their opposition win fairly often as well, more so in cup competitions. But in no world should anyone ever be more concerned about what was the probability of winning the game after you've won, which apparently seems to be a trend these days. And the fact that xG doesn't directly correlate to the probability of winning a match in the first place.
The 07/08 semi final second leg was a very tight game with few chances for either side - 9 shots for Barca and 7 for United with 3 and 2 shots on target respectively. The xG may have even favoured United looking at the highlights, although they’re from the United channel. This is why I said you’d have to look at the xG data to judge these past games in terms of xG because our memories are very unreliable.
xG should generally be used to look and medium- and long-term trends: your overall xG over a season is a reasonable predictor of the team's limits. If at the halfway stage of a season your team's expected points per game stands at something like 1.6 you're very unlikely to win the league. It does have some limited value, with the proper context, in looking at individual matches and analysing the performance levels.
And a coach should definitely look at performance levels, not just results, precisely because there's a lot of unpredictability in football. If Barcelona have 30% possession and 1 shot against a third division side but somehow flukes a 1-0 with the minnows hitting the post four times, you'd expect the coach to recognise the need for improvement and not dismiss it with 'we won, end of story'.
Obviously, this is an extreme example but surely it's always valuable to look beyond results.
This just shows how stupid it is to be wrapped up by xG. This will annoy the few rabid xG fetishists, but it's a woeful way to analyse games on a game by game basis. To look at a general trend over a season, I can see it being useful.
PSG defended that game brilliantly, and nullified City. Other than the Sterling and Bernardo chance, City didn't look like scoring at all.
You've already switched to xPts, when I was talking about xG. It's obviously a nice indicator but in no way does it go far in actually determining the probability of results over the course of a season. So the question isn't whether people are not giving emphasis to the probability of favourable results for a team but the logic of deducing that probability using xG.
You are assuming that the said Barcelona team and coach intend to use tactics where they want to dominate possession, whereas there are plenty of teams who intentionally forego possession and as a result allow the other team to have more shots and find means to consistently win using that tactic. It completely depends on how the team wants to play, and you can increase the probability of winning without necessarily increasing your xG.
Regardless of that the debate isn't about the ability of forecasting results based on xG - which is another debate in itself - but it is around reacting to a result based more on the xG than on the actual result. Going by this forum Poch should be criticized today after getting a 2-0 win against possibly the best team in the tournament - which most would find laughable.
xG has its flaws but last season no team in the PL finished 2 spots off where the xG told them to be except Brighton who were 9 places lower than their xG.
I can appreciate that and that's obviously good evidence in favour of the statistic, but at the same time not a proof if your concept is more xG = more points.
No I agree and yeah it also punishes the better players and teams who will score more times from positions they shouldn’t etc
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/4177524/predict_test_gdpts.0.png)
xG is useful.
Are you saying Chelsea's 2004-06 back to back winning PL teams that were based on the best defence in PL history would have had lower xG than their opponents over the season?
As in, on a game to game basis, they ended up with lower xG than their opponents more often than not? Because that team was built entirely around giving the opposition zero chances, and they were better at it than any other team before or after. I bet their xG differential would have been the highest in the league by a mile in both those seasons (if this stuff was recorded back then, obviously).
As for the general claim that plenty of teams have based entire seasons and won titles on the strategy of having lower xG than the other team, I'd love to see a few examples. Because that sounds... completely wrong.
Last edited:
Was going to write the same question. Chelsea one season conceded 15 goals and I think due to Mourinho career that they get called a parking the bud side yet they scored loads and where attacking most sides
Horrendous.
How else will they ever win the prestigious xG League?
xG is the most useful but most misused stat there is.
xG is useful.
Using it on a game by game basis makes literally zero sense as there simply aren't enough data points to predict the team's fortunes
I’d be interested to know how often the team with the lower xG in a game still wins. I get the feeling it could be more often than one might instinctively think.
If used in the right context over the course of a season.
xG is useful.
Not over one game.
And it's not to be relied upon for anything other than a metric to highlight the quality of chances.
In PL last season, if we classify the games where xG difference of =<0.3 as a draw, the xG results and actual results were same in 212 games (55%) and there was an undeserving winner in 41 games (11%)
If we classify the games where xG difference of =<0.5 as a draw, the xG results and actual results were same in 199 games (52%), and there was an undeserving winner in 29 games (8%).
So, on a game by game basis, xG isn't that great a model like a lot of people have mentioned earlier