Dean Smith - Confirmed

[Dante Lavelli]

Measuring xG is a smart move.  Intangible enough to be impossible to accurately measure yet a good barometer of performance in games.  A run of bad luck/results could mean wholesale changes yet the underlying performances/trend is positive.

it is also slightly otherworldly so allows the coaches to engineer (blind) faith in their approach rather than just the results.

[Steve67]

Sod this xG malarcky, Stick the ball in the net for real and then count it  all up after game! Easy peasy and far more tangible, oh, and measurable.

[paul_e]

The point of xG is to try to give a idea of how fair a reflection of performances the results are. We've all watched games where a team has been battered for 89 minutes and then snatched a late winner, it's numbers to back up that feeling of good/bad luck. I like it because it's normally a pretty fair reflection of the game. For example of our last 4 games it suggests:

-West Ham was pretty even with us having the better chances and they scored a fluke, Antonio missed their best chance
-Arsenal was pretty even but we scored a goal from a very tough chance and they missed a comparative sitter through Nketiah before we closed shop and defended well for the last 15-20mins
-Everton both teams missed a very good chance (El Ghazi and Calvert-Lewin), we scored from our other good chance and they scored one that came from nothing and we should've dealt with
-Palace we had the 3 best chances, scored 2 of them and Samatta missed the other, Palace had a few attempts but nothing you'd call a good chance except a tough one for Van Aarnholt

Does anyone disagree with any of those being a fair reflection of the games?

These stats don't replace watching the game and following play, they just back up that gut feeling we all get.

[eamonn]

I kinda miss FootySkillz. He would weigh-in with an interesting take on this.

[Gareth]

For all this xG nonsense the basics of not being able to defend the area betw the width of the penalty area to the touch line has to be addressed or those crosses raining in will cost Dean his job

[algy]

The point of xG is to try to give a idea of how fair a reflection of performances the results are. We've all watched games where a team has been battered for 89 minutes and then snatched a late winner, it's numbers to back up that feeling of good/bad luck. I like it because it's normally a pretty fair reflection of the game. For example of our last 4 games it suggests:

-West Ham was pretty even with us having the better chances and they scored a fluke, Antonio missed their best chance
-Arsenal was pretty even but we scored a goal from a very tough chance and they missed a comparative sitter through Nketiah before we closed shop and defended well for the last 15-20mins
-Everton both teams missed a very good chance (El Ghazi and Calvert-Lewin), we scored from our other good chance and they scored one that came from nothing and we should've dealt with
-Palace we had the 3 best chances, scored 2 of them and Samatta missed the other, Palace had a few attempts but nothing you'd call a good chance except a tough one for Van Aarnholt

Does anyone disagree with any of those being a fair reflection of the games?

These stats don't replace watching the game and following play, they just back up that gut feeling we all get.

That's pretty much my view too, along with what SE said about quite possibly using other datasets and that xG was just the one people are most familiar with/he was willing to let on that we put focus on.

Not saying it should be the be-all-and-end-all, but i think we should absolutely be using statistical analysis to improve as many areas of the club as we can. E.g. xG could reasonably be used to measure if we're creating better chances for the forwards, even if they're not scoring those. Or if we're restricting the opposition to more difficult shots for that matter. We should also be able to measure how effective stuff like fitness training is, both if the players can keep going for 90+ mins, but also if we've got the right balance of sprints vs jogging for each position.

I suppose where i'm going is that the focus for sort of stats should be to measure the effectiveness of training rather than absolute performance of the team.

[Fred Crump]

I kinda miss FootySkillz. He would weigh-in with an interesting take on this.

I miss Footy/ Vill I  an too, is he taking one of his sabbaticals again? He could always be relied upon to come at things from an ‘alternative’ viewpoint, which is surely what a Football chat forum should be about, I guess. I get the impression that he was a bit of a sensitive soul and sometimes was hurt by the stick we gave him. If so, don’t take it too seriously Footy , and please come back soon.

[eamonn]

Yeah, I think he said he'd be back in September.

[Damo70]

Sod this xG malarcky, Stick the ball in the net for real and then count it  all up after game! Easy peasy and far more tangible, oh, and measurable.

I miss the good old days of Alan Hansen telling me about "shocking" defending and how you won't win anything with kids.

[Sdwbvf]

Don't Liverpool have a team of 20 odd statisticians? It doesn't seem to have done them any harm.

[AsTallAsLions]

Don't Liverpool have a team of 20 odd statisticians? It doesn't seem to have done them any harm.

We just have one. His name is Footyskillz

[charlatan]

Don't Liverpool have a team of 20 odd statisticians? It doesn't seem to have done them any harm.

Apparently last season they managed to outperform their xGD by a net 15 goals and bag 25 extra points as a result.
https://understat.com/league/EPL

[AsTallAsLions]

🙄

[Sdwbvf]

Don't Liverpool have a team of 20 odd statisticians? It doesn't seem to have done them any harm.

Apparently last season they managed to outperform their xGD by a net 15 goals and bag 25 extra points as a result.
https://understat.com/league/EPL

And we underperformed by 5 points

[paul_e]

The most interesting thing about that is that it backs up the people who think Bruce is a fucking shit manager.