Statistical Domination

[hogesar 9,658]

Likely won't surprise people that xG shows we should have won the game having created the much better opportunities in the game. Actually slightly surprised Pukkis xG isn't a bit higher.

[norfolkngood 1,086]

5 minutes ago, hogesar said:

Likely won't surprise people that xG shows we should have won the game having created the much better opportunities in the game. Actually slightly surprised Pukkis xG isn't a bit higher.

i do not like stats  WBA have the highest XG in the league at home and  2nd overall and they are 3rd from bottom 

[chicken 2,624]

On 07/10/2022 at 15:39, Shining White said:

Hugely unfair? Then tell me which player/s from all of the teams we have played so far would you swap for one of our first team?

This could be a nice one to answer.

I mean you can rattle off the likes of Sarr, who is a cut above the vast majority of players at this level and would just instantly demote one of our wide players. Berge who could walk into our CM. Egan would probably get the nod at CB. 

I actually think it's easier to ask which of our players would walk into one of their starting 11's... Pukki, Sargent... then it becomes a bit murky. We all love Gibbs but he's young and other teams have more experienced players in that position so would unlikely pick him ahead of a more consistent older player. Aarons probably would, but perhaps isn't quite the shoe in you would have put him down to be a couple of seasons ago. Krul? Not on yesterdays performance. Nunez? - again, not yet, other teams have players in that position performing consistently, even an "on paper" argument holds no ground here when compared to an experienced player who has played premier league football. Hanley is another possible/outside.

So really just Pukki and Sargent. That's it for me.

I think people value Gibbs and Omobamidele for their potential and how good they are at this stage in their young careers. In reality though, are they better than other players out there who can at least be consistently better performers if not quite so good on their day? 

[Robert N. LiM 4,340]

39 minutes ago, hogesar said:

Likely won't surprise people that xG shows we should have won the game having created the much better opportunities in the game.

xG has it as our most convincing performance of the season, going by the graphs you've collected at the start of the thread.

[chicken 2,624]

21 minutes ago, norfolkngood said:

i do not like stats  WBA have the highest XG in the league at home and  2nd overall and they are 3rd from bottom 

Only one half of the story though isn't it?

Scored 17 goals, joint with Swansea in 6th, better than Reading on 15 in 5th. Also more than Blackburn in 7th, Preston in 8th... In fact, 17 'goals for' is better than 15 teams and equal to a further three teams with only 5 teams having scored more.

Defensively they have conceded 18 goals, only one team in the top half of the table has managed that - Reading. Otherwise you have to go down to 18th and Bristol City who have conceded 21. In fact, only three teams have conceded more than 18 goals. Weirdly, none of them are currently in the bottom 3. Hull have conceded the most - 26, but sit in 20th 3pts off WBA in 22nd.

The other factor here is how those goals are conceded/scored across games. A 6-0 win gives you an amazing goal difference, but if that was the only win in 6 games and the others being 0-0 draws, you'd have 8pts, another team might go W:2-0, L:1-0, W:2-1, D:1-1, L:1-0, W:1-0 and have a total of 6 goals for, 4 goals against but have 10pts having scored the same number of goals, and let more goals in. 

[nutty nigel 7,531]

Genuine question...

What does xg suggest about Reading v WBA next week (5th at home to 22nd)?

For those who think it's a load of old pony you would think it would be odds on a home win. And yet home win is 9/4, draw is 11/5, away win is 23/20. 

Edited October 9, 2022 by nutty nigel

[Midlands Yellow 3,980]

15 minutes ago, nutty nigel said:

Genuine question...

What does xg suggest about Reading v WBA next week (5th at home to 22nd)?

For those who think it's a load of old pony you would think it would be odds on a home win. And yet home win is 9/4, draw is 11/5, away win is 23/20. 

Nutty!! For the first time ever I love your question.
There’s something in Xg no doubt but ffs let’s see some predictions until the end of the season. You win, charity wins, and so do individuals if there’s anything in it. 

[nutty nigel 7,531]

16 minutes ago, Midlands Yellow said:

Nutty!! For the first time ever I love your question.
There’s something in Xg no doubt but ffs let’s see some predictions until the end of the season. You win, charity wins, and so do individuals if there’s anything in it. 

I don't understand it enough to answer questions.

I looked at the Reading WBA game and 9/4 for a home win 5th v 22nd looks huge.

Is that xg influence? Because if so, and xg is nonsense, we should fill our boots!

Edited October 9, 2022 by nutty nigel

[repman 635]

59 minutes ago, nutty nigel said:

I don't understand it enough to answer questions.

I looked at the Reading WBA game and 9/4 for a home win 5th v 22nd looks huge.

Is that xg influence? Because if so, and xg is nonsense, we should fill our boots!

As far as I can tell, WBA's xG is inflated by 2 facts.

1. Their keeper isn't very good, if you consider that an xG of 0.33 for a shot is probably what most people would consider a good chance, then if your keeper is letting them in you're already down -0.67 from your goals conceded to xG conceded.

2. With the keeper being bad (or just letting everything in right now), WBA have been behind a lot, which has meant they will be more likely to have more shots and concede less shots.

Reading on the other hand have been pretty good at getting the lead and then seeing it through. As long as they're 1-0 up they are happy enough to sit on the lead, which would mean their xG difference possibly looks worse.

[hogesar 9,658]

1 hour ago, nutty nigel said:

I don't understand it enough to answer questions.

I looked at the Reading WBA game and 9/4 for a home win 5th v 22nd looks huge.

Is that xg influence? Because if so, and xg is nonsense, we should fill our boots!

Yes, although typically before xG was mainstream you'd have got significantly better odds on WBA, as their 22nd position would have been enough to assume a hugely likely loss for them. As it is, bookies now know via the collected data just how close they've been to winning the majority of their games.

[hogesar 9,658]

2 hours ago, Robert N. LiM said:

xG has it as our most convincing performance of the season, going by the graphs you've collected at the start of the thread.

Yeah although I would like to know why one of Pukkis chances aren't rated higher and why Prestons second goal didn't have a higher ranking either. But like most of us, I'm still learning 🤓

[canarybubbles 1,915]

I accept there are Luddites like me who aren't very interested in things like xG and who should perhaps drag ourselves into the 21st century. However, there are also the opposite - people who are mesmerised by numbers (it's science, innit?) and never question how the criteria behind the numbers are chosen. 

My biggest concern, though, is that with so much data around nowadays, people often just go scouring to find data which fits their argument, the very opposite of a scientific use of statistics.

OK, I'm a dinosaur, but I still prefer the eye test if we have to choose between that and data (of course, we don't have to, but it seems like with everything these days, we have to polarise into two opposing camps).

[littleyellowbirdie 2,547]

1 hour ago, canarybubbles said:

I accept there are Luddites like me who aren't very interested in things like xG and who should perhaps drag ourselves into the 21st century. However, there are also the opposite - people who are mesmerised by numbers (it's science, innit?) and never question how the criteria behind the numbers are chosen. 

My biggest concern, though, is that with so much data around nowadays, people often just go scouring to find data which fits their argument, the very opposite of a scientific use of statistics.

OK, I'm a dinosaur, but I still prefer the eye test if we have to choose between that and data (of course, we don't have to, but it seems like with everything these days, we have to polarise into two opposing camps).

Almost all discussion of football seems to reduce to 'should we sack the manager?', which is a binary thing, so inevitably will polarise. Personally, I would love to see managerial hiring and firing restricted to the Summer only, ideally; managerial turnover throughout the sport is becoming absurd as ever more desperate clubs look to appease ever more demanding fans with the only relatively easy change at their disposal that can happen any time, as often as you like. 

Edited October 10, 2022 by littleyellowbirdie

[CANARYKING 637]

We lost, move on

[Shining White 13]

12 hours ago, chicken said:

This could be a nice one to answer.

I mean you can rattle off the likes of Sarr, who is a cut above the vast majority of players at this level and would just instantly demote one of our wide players. Berge who could walk into our CM. Egan would probably get the nod at CB. 

I actually think it's easier to ask which of our players would walk into one of their starting 11's... Pukki, Sargent... then it becomes a bit murky. We all love Gibbs but he's young and other teams have more experienced players in that position so would unlikely pick him ahead of a more consistent older player. Aarons probably would, but perhaps isn't quite the shoe in you would have put him down to be a couple of seasons ago. Krul? Not on yesterdays performance. Nunez? - again, not yet, other teams have players in that position performing consistently, even an "on paper" argument holds no ground here when compared to an experienced player who has played premier league football. Hanley is another possible/outside.

So really just Pukki and Sargent. That's it for me.

I think people value Gibbs and Omobamidele for their potential and how good they are at this stage in their young careers. In reality though, are they better than other players out there who can at least be consistently better performers if not quite so good on their day? 

The question was clear - which players OF THE TEAMS WE HAVE PLAYED SO FAR would you swap? i.e. we have had the better squads up to now in our fixtures, and you reply with a player from Watford and 2 from Sheffield United??!! 🤦‍♂️🤣

[Jim Smith 2,317]

12 hours ago, hogesar said:

Likely won't surprise people that xG shows we should have won the game having created the much better opportunities in the game. Actually slightly surprised Pukkis xG isn't a bit higher.

Surprised Ramsey's first half chance after Kenny bundled his way through is so low as well. 5 yards out. harder not to score that one. 

[Jim Smith 2,317]

11 hours ago, nutty nigel said:

I don't understand it enough to answer questions.

I looked at the Reading WBA game and 9/4 for a home win 5th v 22nd looks huge.

Is that xg influence? Because if so, and xg is nonsense, we should fill our boots!

The WBA situation is interesting isn't it. With us, I took our xG stats earlier in the season as evidence that if we kept doing what we were doing then things would pick up and that is precisely what has happened.

With WBA though it hasn't done and the fact its now gone on so long does tend to suggest that they have certain issues that have not been resolved. Probably a below average keeper and strikers.

[hogesar 9,658]

13 minutes ago, Jim Smith said:

Surprised Ramsey's first half chance after Kenny bundled his way through is so low as well. 5 yards out. harder not to score that one. 

I'm not 100%, but as xG is the chance from the point the player shoots, at that point there were defenders either side and the goalkeeper had come out quite well and closed down the space. Still a good opportunity but I guess they relatively frequently get blocked / saved under that sort of pressure. 

Again, I've not seen replays so I'm just going by memory, although it was right in front of me as i'm lower barclay.

[Ken Hairy 3,772]

11 hours ago, nutty nigel said:

I don't understand it enough to answer questions.

I looked at the Reading WBA game and 9/4 for a home win 5th v 22nd looks huge.

Is that xg influence? Because if so, and xg is nonsense, we should fill our boots!

So what now? Go with xG or the nmB (new manager Bounce). Lump on the draw??? 😏

[wcorkcanary 4,334]

Someone messed up spectacularly with Haalands XG!! To allow  @Feedthewolf to get 25/1 on him scoring 3 hattricks in the whole season then . 😇

[Robert N. LiM 4,340]

3 hours ago, canarybubbles said:

However, there are also the opposite - people who are mesmerised by numbers (it's science, innit?) and never question how the criteria behind the numbers are chosen. 

Good post, though I would say that on this thread there are lots of examples of stats enthusiasts digging into the data to try and understand those criteria.

My view is that you should use both stats and what you call 'the eye test' - using the stats to confirm your gut feeling (which is always liable to bias, especially when it's the club that you support), but being critical of the stats and understanding their limitations. I think that quite a lot of people think it's one or the other, but I'd strongly dispute that.

[Badger 2,405]

3 hours ago, canarybubbles said:

I accept there are Luddites like me who aren't very interested in things like xG and who should perhaps drag ourselves into the 21st century. However, there are also the opposite - people who are mesmerised by numbers (it's science, innit?) and never question how the criteria behind the numbers are chosen. 

My biggest concern, though, is that with so much data around nowadays, people often just go scouring to find data which fits their argument, the very opposite of a scientific use of statistics.

OK, I'm a dinosaur, but I still prefer the eye test if we have to choose between that and data (of course, we don't have to, but it seems like with everything these days, we have to polarise into two opposing camps).

Nothing wrong with using "the eye test" as well! The eye test for Saturday is that we missed some very good chances and gave away some soft goals - which is pretty much what the data says. I'm not a WBA fan, but my guess is that their fans will be bemoaning "soft goals" that the team gave away and missed chances, again supported by the data - I saw the game on TV on Friday and actually felt sorry for Steve Bruce!

You are entirely correct about people picking statistics that suit their arguments but in such case it is the misuse of statistics that is the issue that is the problem not the statistics themselves - most obviously done in politics. To use statistics properly you need to look at all the data and also be aware of its limitations.

 

[king canary 7,605]

26 minutes ago, Robert N. LiM said:

Good post, though I would say that on this thread there are lots of examples of stats enthusiasts digging into the data to try and understand those criteria.

My view is that you should use both stats and what you call 'the eye test' - using the stats to confirm your gut feeling (which is always liable to bias, especially when it's the club that you support), but being critical of the stats and understanding their limitations. I think that quite a lot of people think it's one or the other, but I'd strongly dispute that.

Yes I remember there were a couple on here who swore blind that they were able to give their opinions on player performances without having seen the game but by looking at heat maps and stats which is just nonsense in my view. 

[hogesar 9,658]

Just now, king canary said:

Yes I remember there were a couple on here who swore blind that they were able to give their opinions on player performances without having seen the game but by looking at heat maps and stats which is just nonsense in my view. 

To be fair though, you're on a forum where a previous poster genuinely thought he could analyse the game better than most from the comfort of his living room because he had a bigger tv than most so could see more of the pitch...

[nutty nigel 7,531]

2 hours ago, Ken Hairy said:

So what now? Go with xG or the nmB (new manager Bounce). Lump on the draw??? 😏

Lol 

We looked into the new manager bounce a few years ago. It's a myth. It's something we notice when it happens but don't when it doesn't.

Currently Reading 12/5, draw 9/4, WBA 11/10

 

 

 

[Robert N. LiM 4,340]

49 minutes ago, king canary said:

Yes I remember there were a couple on here who swore blind that they were able to give their opinions on player performances without having seen the game but by looking at heat maps and stats which is just nonsense in my view. 

Didn't know the CEOs of Brentford and Brighton posted on this board. 😉

[Robert N. LiM 4,340]

48 minutes ago, hogesar said:

To be fair though, you're on a forum where a previous poster genuinely thought he could analyse the game better than most from the comfort of his living room because he had a bigger tv than most so could see more of the pitch...

That's obviously ridiculous, but when I'm watching a dodgy stream on my (reasonably small) laptop, I often find it difficult to tell the players apart, so find it harder to say with confidence who I think is having a good game. I thought Sara was excellent on Saturday, but I'm now wondering how much that's to do with his enormous, instantly recognisable hair. 

[nutty nigel 7,531]

2 hours ago, Jim Smith said:

The WBA situation is interesting isn't it. With us, I took our xG stats earlier in the season as evidence that if we kept doing what we were doing then things would pick up and that is precisely what has happened.

With WBA though it hasn't done and the fact its now gone on so long does tend to suggest that they have certain issues that have not been resolved. Probably a below average keeper and strikers.

So is xg difference better? 

Just looking at a site called infogol...

Reading would be -0.4 where as WBA would be +7.9. But then should we break that down to home and away?

[nutty nigel 7,531]

3 hours ago, CANARYKING said:

We lost, move on

No

[Paint Me Yellow 200]

Got to love a bit of spreadsheet football after a $h!t-shower of a game. Can't wait for the slide-show to follow.

It'll only have one slide with "But...but...2nd" written on it.