| xG table 18:37 - Nov 6 with 1034 views | rkc123 |
I am not hugely into xG and those sort of statistics, but having been told by a friend that we are 2nd in the xG table I thought I would have a look, and out of interest did a comparison to where we were at the same stage in the 23/24 season.
Interestingly having had our game against Rotherham postponed that year we had only played 12 games when the most of the rest of the league had played 13, so it is a like for like comparison with this season. We were 2nd on 31 points, but were 5th at the time on the xG table (expected points of 21), we had scored 26, 4 more than our xG, and had conceded 13, 2 less than xGa. This season we have 19 points (3 fewer than expected), and have scored 2 goals less than our xG of 23.1, so we actually currently have a higher xG and expected points at this point of the season than we did in 2023.
That's a lot of numbers, but I think it sort of demonstrates the feeling most supporters have now (and had back then). In 23/24 I remember being at games just feeling like we were going to win, we would get that late goal, the close games were going our way, we would take that one chance. This season on the other hand it just feels like things aren't quite going our way, we aren't taking the chances, we are losing games by the odd goal rather than winning them, and the crowd don't have that confidence in the team.
I think the atmosphere in the ground late on against Watford was noticeably different to what it would have been in a similar match situation in 23/24, it didn't feel in the ground like the crowd expected us to win the game, and there wasn't that ratcheting up of intensity from the fans or players.
I am not sure which comes first, the crowd bringing that positive energy towards the end of close games or a couple of close games going our way to get fans believing again, but I think if we are to get on a roll this season then late goals are going to be as important as in our last promotion season - particularly as we have more depth and quality off the bench than most teams in the league so in theory the disparity in quality between us and our opposition will be at its greatest after each team have made their substitutions. [Post edited 6 Nov 22:02]
|  | | |  |
| xG table on 19:06 - Nov 6 with 918 views | swede |
We are bombarded with statistics every match these days, percentage possession, shots on target/off target, completed passes, tackles, corners etc etc.
I have still to read an explanation of xG that makes the slightest bit of sense to me.
Back in the day the only statistics that counted were 1) Goals scored, 2) Goals conceded. That has not changed. | | | |
| xG table on 19:11 - Nov 6 with 892 views | BrianTablet |
| xG table on 19:06 - Nov 6 by swede |
We are bombarded with statistics every match these days, percentage possession, shots on target/off target, completed passes, tackles, corners etc etc.
I have still to read an explanation of xG that makes the slightest bit of sense to me.
Back in the day the only statistics that counted were 1) Goals scored, 2) Goals conceded. That has not changed. |
It's just quantifying things like 'we missed a hatful' or 'he missed a sitter' or 'on the day we couldn't hit a cow's arse with a banjo' |  |
| |
| xG table on 19:45 - Nov 6 with 759 views | stonojnr |
| xG table on 19:11 - Nov 6 by BrianTablet |
It's just quantifying things like 'we missed a hatful' or 'he missed a sitter' or 'on the day we couldn't hit a cow's arse with a banjo' |
But it doesn't though our xG was higher than Charltons was, they scored 3, we scored 0, nobody left that game thinking we missed a hatful or any real sitters We could have still been playing that game and we wouldn't have scored yet. | | | |
| xG table on 19:47 - Nov 6 with 749 views | TNBlue |
| xG table on 19:06 - Nov 6 by swede |
We are bombarded with statistics every match these days, percentage possession, shots on target/off target, completed passes, tackles, corners etc etc.
I have still to read an explanation of xG that makes the slightest bit of sense to me.
Back in the day the only statistics that counted were 1) Goals scored, 2) Goals conceded. That has not changed. |
I hate football stats. The only stat that matters is getting the ball in the net. | | | |
| xG table on 20:05 - Nov 6 with 685 views | Swansea_Blue |
| xG table on 19:11 - Nov 6 by BrianTablet |
It's just quantifying things like 'we missed a hatful' or 'he missed a sitter' or 'on the day we couldn't hit a cow's arse with a banjo' |
Yep. It’s interesting but just needs to be taken in context. It’s a stat that helps you understand but doesn’t prove anything. It doesn’t matter if you have a rubbish xG if you’ve got someone who can score worldies each week. Similarly, it can reinforce the view that you’re maybe not as bad as results suggest if you’re falling your xG (which is where we are) but there’s little comfort in that as goals are all that matter. Like any stat, it’s only a tool that’s useful when analysed with understanding. |  |
| |
| xG table on 20:12 - Nov 6 with 655 views | Guthrum |
| xG table on 19:06 - Nov 6 by swede |
We are bombarded with statistics every match these days, percentage possession, shots on target/off target, completed passes, tackles, corners etc etc.
I have still to read an explanation of xG that makes the slightest bit of sense to me.
Back in the day the only statistics that counted were 1) Goals scored, 2) Goals conceded. That has not changed. |
Think of it like launching a rocket into space.
The only statistic which really matters is whether it makes it to orbit or not. Horizontal speed and altitude.
However, a lot of other stats will be feeding into that, informing decisions, monitoring progress. Engine chamber pressure, thrust-to-weight ratio, fuel consumption, angle of trajectory, dynamic pressure and many, many more. All looking at aspects which are important to the ultimate failure or success of the launch.
They are all useful to those running the project. What things to tweak, what design directions to take, highlighting what has gone wrong if it blows up or crashes.
Winning a football game is neither easy nor simple at a high level. There are so many variables - team selection, setup, tactics, form, morale, effects of recent matches, attitude of the fans, pressure upon the staff, all of those same aspects relating to the opposition, the Ref, even weather and which way the wind is blowing or sun shining.
The more angles coaches have to look at everything, the better they can identify advantages and avoid errors. You can get so far on managerial gut instinct, but using it alone in this day and age will set you behind your rivals. |  |
| |
| xG table on 20:24 - Nov 6 with 613 views | bartyg |
| xG table on 19:06 - Nov 6 by swede |
We are bombarded with statistics every match these days, percentage possession, shots on target/off target, completed passes, tackles, corners etc etc.
I have still to read an explanation of xG that makes the slightest bit of sense to me.
Back in the day the only statistics that counted were 1) Goals scored, 2) Goals conceded. That has not changed. |
old man yells at cloud [Post edited 6 Nov 20:24]
|  | | |
| xG table on 20:26 - Nov 6 with 597 views | Herbivore |
| xG table on 19:45 - Nov 6 by stonojnr |
But it doesn't though our xG was higher than Charltons was, they scored 3, we scored 0, nobody left that game thinking we missed a hatful or any real sitters We could have still been playing that game and we wouldn't have scored yet. |
Akpom missed two sitters alone. We missed other decent chances too. |  |
| |
Login to get fewer ads
| xG table on 20:32 - Nov 6 with 574 views | BrianTablet |
| xG table on 20:26 - Nov 6 by Herbivore |
Akpom missed two sitters alone. We missed other decent chances too. |
And therein lies it's point I guess - an objective measure of fact, not a subjective flawed memory of events. | |
| |
| xG table on 20:54 - Nov 6 with 520 views | rkc123 |
| xG table on 20:05 - Nov 6 by Swansea_Blue |
Yep. It’s interesting but just needs to be taken in context. It’s a stat that helps you understand but doesn’t prove anything. It doesn’t matter if you have a rubbish xG if you’ve got someone who can score worldies each week. Similarly, it can reinforce the view that you’re maybe not as bad as results suggest if you’re falling your xG (which is where we are) but there’s little comfort in that as goals are all that matter. Like any stat, it’s only a tool that’s useful when analysed with understanding. |
Yes exactly, often teams 'regress to the mean' so to speak, but that is not always the case. In the 23/24 season fans of other teams near the top would often use the fact we out performed our xG and expected points as evidence that we were 'lucky', but you don't get lucky over a whole season, obviously there was more to it than that. I brought up the stats I did though as I do think it shows that the margins have been fine, I don't think we are playing anything close to as well as we can with the quality we have in the team, and yet if we took a few more of our chances we could be several points better off. I am hopeful that if we can take a few more of the chances we are creating and string a couple of results together then we can start to build that confidence amongst the team and supporters again, and I really think there's the potential for us to get on quite a roll during the busy run of fixtures coming up if that happens. | | | |
| xG table on 20:56 - Nov 6 with 518 views | Garv |
I actually disagree regarding the Watford game. There was a 5-10 minute spell, probably around 75-80, where I did get a similar feeling to the one that we so often had late in games during 'that' season.
That could be taken both ways. Maybe we're getting back to those times slowly, or we have a good team with as much or more quality but just don't have the drive or team spirit to get wins over the line. |  |
| |
| xG table on 20:56 - Nov 6 with 515 views | rkc123 |
| xG table on 19:45 - Nov 6 by stonojnr |
But it doesn't though our xG was higher than Charltons was, they scored 3, we scored 0, nobody left that game thinking we missed a hatful or any real sitters We could have still been playing that game and we wouldn't have scored yet. |
I agree by the end we could have been playing all night and wouldn't have scored, but we should have been ahead at half time, and I think if we get the lead in that game we would have won it. Obviously the reaction to going behind was the most concerning thing, and that is the sort of thing that you don't see in statistics. | | | |
| xG table on 21:01 - Nov 6 with 491 views | Smoresy |
Proves we've created a good amount more than we've faced over these 13 games, which I'm confident we've mostly seen with our own eyes. Continue doing that and my hope remains high that we'll end up alright. We don't need to be exceptional in front of goal if we dominate the chances, just reach average status. | | | |
| xG table on 21:04 - Nov 6 with 485 views | rkc123 |
| xG table on 20:56 - Nov 6 by Garv |
I actually disagree regarding the Watford game. There was a 5-10 minute spell, probably around 75-80, where I did get a similar feeling to the one that we so often had late in games during 'that' season.
That could be taken both ways. Maybe we're getting back to those times slowly, or we have a good team with as much or more quality but just don't have the drive or team spirit to get wins over the line. |
Yes I think we had a good spell and it did build, but then the intensity dropped off again (both in the players and the crowd), and despite only needing a goal to win I thought the atmosphere was quite flat and tense from 80 minutes on, sort of just waiting for something to happen. | | | |
| xG table on 08:01 - Nov 7 with 185 views | bsw72 |
xG doesn't get you points therefore won't get you promoted, no matter how "good" it indicates we may well be playing / creating chances. [Post edited 7 Nov 8:03]
|  | | |
| xG table on 08:07 - Nov 7 with 171 views | Herbivore |
| xG table on 08:01 - Nov 7 by bsw72 |
xG doesn't get you points therefore won't get you promoted, no matter how "good" it indicates we may well be playing / creating chances. [Post edited 7 Nov 8:03]
|
Has anyone ever suggested that xG gets you points? I find it weird when people say that because I don't think I've ever heard anyone suggest anything approaching the idea that you should get points for your xG. | |
| |
| xG table on 08:15 - Nov 7 with 158 views | SuffolkPunchFC |
| xG table on 08:01 - Nov 7 by bsw72 |
xG doesn't get you points therefore won't get you promoted, no matter how "good" it indicates we may well be playing / creating chances. [Post edited 7 Nov 8:03]
|
That's somewhat stating the obvious, and demonstrates a lack of understanding of the point of stats such as xG, and how they should be used. | | | |
| xG table on 08:46 - Nov 7 with 118 views | bsw72 |
| xG table on 08:15 - Nov 7 by SuffolkPunchFC |
That's somewhat stating the obvious, and demonstrates a lack of understanding of the point of stats such as xG, and how they should be used. |
OK, calm down. I do understand statistics and their applied use as it forms a large part of my work. My initial reply was simply addressing someone referencing xG as a standalone measure of performance with a throwaway comment.
I’m happy to discuss xG stats in detail, but it’s crucial to clarify which model is being used, basic or advanced. At a very basic level at least reference expected goals against (xGA) and its inverse relationship with xG, which would add a little depth to the analysis.
Secondly specifying the xG model and without referencing other football statistics for context, relying on xG alone is essentially meaningless. xG is not a universal constant; its value and interpretation vary significantly depending on the model’s design, data inputs, and assumptions. Ignoring this nuance reduces the metric to an empty number, detached from the broader analytical framework that gives it relevance and insight.
I’m always open to debating the effectiveness of statistics in football analysis and wider. I had a similar discussion this summer with the head of data science at McLaren, who argued that the human element and the many variances in football make statistical analysis fundamentally different and arguably more challenging compared to Formula 1, and therefore potentially less use. | | | |
| |
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
