General Premier League Thread 2023-24 Mod Note in op 27/6/23 And 21/05/24

suspended. turns out if a keeper gets a 2nd yellow in pen's he does not get sent off but suspended for the next game

Imo it was half a dozen of one against the other and harsh to call it as a foul.

Ah it’s definitely a foul. You would expect it to be given in real time, but I’m surprised they overturned it on VAR

For xG do they not break the pitch down into smaller sections and then analyse how many shots have been taken from that area and how many goals were scored from those shots?

It's obviously only a rough guide in general but it does have some merit.

xg is not useful at all for analysing a one off game, unless for people who can't judge a game by watching it.

actually that’s an interesting one, cause he didn’t - it looked like he did at first glance, with a goal kick given after the ‘shot’ went wide, but replays showed the defender got a great foot in and cleared it before it ever even reached Nunez - so in that instance did the xg record the incident the way it was reffed rather than what actually happened? Or maybe it’s one that gets corrected later.

I had to look it up. It’s the one he got on target. Once he’s in that position it has to be scored https://youtu.be/J8pwU2yZZuc&t=84

xG I wasn’t sure about till I saw a post saying basically a penalty is an xG of .71, basically 71 of every 100 penalties are scored, so that makes sense. It’s obviously more accurate the more data they have for every situation. A one off game isn’t much use, so like most things with data the more data samples you have the more you can tell from it, over the course of a season a player exceeding his xG he’s worth a look at, a player underperforming it is probably not worth looking at is how clubs would look at it perhaps.

Well it definitely wasn't by using a shots on target stat anyways.

I don't think clubs ever used assists as a metric, and if they did they should be shot. And then even more so if they are using assists to judge a central midfielder.

There are other stats such as xGchain which can record midfielders contribution in terms of creating chances consistently which are helpful when analyzing midfielders creativity. It's one of many stats what are useful.

It's funny when you look at the 'old school' stats and you realize how useless they are really. There was one going around a few years who where some Man Utd fans used to love parroting the fact that David De Gea had made the most saves in the PL, to show what a great goalkeeper he was. First of all if a goalkeeper is making a lot of saves, that means he is facing a lot of saves and that is not a good thing. And then second of all, deeper stats were showing that while making saves, he was not catching them and instead was rebounding the saves into dangerous areas which in turn was creating another chance for the opposition. In layman terms, he was saving shots from distance (typically harder to score from) but was palming them into the 6 yard box for an easier chance for the opposition to score from a tap in . The old school metric of 'shots saved' was painting this picture that he was great as he was making saves but in reality he helping the opposition with where he was putting the ball and so the new metric of psxg was measuring his shot saving performances more accurately. He was severely underperforming.

They definitely use it, yes. And stats like xg, and it's dependent stats, are a large reason why some clubs have gained an advantage over others when recruiting new players and can exploit gaps in the market. And then you can see the clubs who are spending big money and being left behind as they are buying the wrong players.

I was watching a game recently, it might have been the Chelsea cup semi final and a player ran through on goal one on one with the keeper, rounded him and defenders got back and he didn't even take a shot. So that didn't even exist according to xg.

If a player doesn't take a shot, then how do you expect it to be registered as any type of shot stat?

I think people just struggle to comprehend that the idea of a perfect stat not being possible to predict future outcomes, doesn’t mean that there are stats available that are far more predictive of future outcomes than looking at previous outcomes.

Like, I would think that Darwin scoring a hatful at some stage is a quite likely thing. But he will still probably always have a lower conversion rate than clinical finishers

I think Arsenal will be too good for Spurs. They have matured a lot since last year even though it may not be enough.

whats the advantage rule these days? cos spurs got zero advantage there

At any point they could've stopped and the ref would have gladly given them a free. They chose to try play on. That's not down to the ref.

That was a lot closer than I thought in real time

goal. hoijberg OG

spurs looking dangerous here from set pieces

this looks very close

offside. sooner they get the automated offside the better