Match Highlights

[PHILINFRANCE]

6 hours ago, Smokey said:

The quick roll out by Bents was great. Something I hope we see more of

It was, although I am not 100% sure he didn’t just venture out of his area slightly before releasing the ball.

On another subject, Andreas Weimann’s excellent second goal, our fifth, has been described as a lob over the keeper.

Surely, it was a chip.

Perhaps the terminology has changed, but when I was playing (many years ago) a lob was the name given to the the sort of shot taken when the ball was bouncing and you just sort of helped it on its way.

[MJP3]

11 minutes ago, 1960maaan said:

His distribution , or lack of it, was why Max got his run. Bents' needed to improve and has, had a hand   in a couple of goals in the last couple of games. 

Not sure about that.

His distribution was great when he first arrived, always looking to start a quick counter.
I assumed the constant long punts that were new to him, were either instructions from the manager or pressure from team mates.

[Smokey]

1 hour ago, Davefevs said:

Dan White, usually does the womens.

I thought he was better.

[Lrrr]

On 30/04/2022 at 18:03, cityexile said:

I would have to resign if I got caught watching this in Parliament. 

You wouldn't want to pre-empt the inquiry

[cidercity1987]

Three one on ones

One open net tap in

Weimanns a beauty but still absolutely clear and unmarked 

Other chances, Martin on the turn 8 yards out, Semenyo cutting in 10 yards out

Expected goals? 1.1

Someone will have to explain that to me as I really don't understand it

[cityexile]

1 hour ago, Lrrr said:

You wouldn't want to pre-empt the inquiry

I was just searching ‘I drove my tractor through your haystack last night...’ Well that, and ‘OnlyFarms’.

Edited May 1, 2022 by cityexile

[Davefevs]

3 hours ago, cidercity1987 said:

Three one on ones

One open net tap in

Weimanns a beauty but still absolutely clear and unmarked 

Other chances, Martin on the turn 8 yards out, Semenyo cutting in 10 yards out

Expected goals? 1.1

Someone will have to explain that to me as I really don't understand it

Not sure the original source of xG, but…

 

[ExiledAjax]

4 hours ago, Davefevs said:

Not sure the original source of xG, but…

 

E361 gave us 1.1. Average of the 4 I use (which includes infogol and E361) gives 1.68 v Hull's 0.79.

E361 has been the stingiest of the 4 I use this season. He gives use a cumulative xG of just 53, or 1.18 per game. My other 3 sources give 59, 57.1, and 56.54.

So my seasonal average of those totals is 56.38. 1.25 per game.

And @cidercity1987 one on one's won't get a huge xG rating. Always remember that a penalty gets only about 0.8, as roughly 80% of penalties are scored. So most other chances will be well below 0.8. Even a one on one will get well below that. Compare it to a penalty - the player is running, ball is moving, goalie can close the angle, maybe a defender is nearby. All of that makes it much less likely to be scored and so brings the xG down to likely 0.3 or something tops.

Edited May 2, 2022 by ExiledAjax

[cidercity1987]

1 hour ago, ExiledAjax said:

E361 gave us 1.1. Average of the 4 I use (which includes infogol and E361) gives 1.68 v Hull's 0.79.

E361 has been the stingiest of the 4 I use this season. He gives use a cumulative xG of just 53, or 1.18 per game. My other 3 sources give 59, 57.1, and 56.54.

So my seasonal average of those totals is 56.38. 1.25 per game.

And @cidercity1987 one on one's won't get a huge xG rating. Always remember that a penalty gets only about 0.8, as roughly 80% of penalties are scored. So most other chances will be well below 0.8. Even a one on one will get well below that. Compare it to a penalty - the player is running, ball is moving, goalie can close the angle, maybe a defender is nearby. All of that makes it much less likely to be scored and so brings the xG down to likely 0.3 or something tops.

 

6 hours ago, Davefevs said:

Not sure the original source of xG, but…

 

2 v 0.6 is more representative of what I saw but that E361 had it at 1.1 v 0.9 with Martins tap in about 0.3 chance. If our chances were a low possibility how did Hull reach 0.9? There was a header in the first half, a blaze over from a cut back in the second and sod all else

I just don't understand the systems and how they rate it all. Martins tap in was about the same chance as pen imo

[ExiledAjax]

4 minutes ago, cidercity1987 said:

 

2 v 0.6 is more representative of what I saw but that E361 had it at 1.1 v 0.9 with Martins tap in about 0.3 chance. If our chances were a low possibility how did Hull reach 0.9? There was a header in the first half, a blaze over from a cut back in the second and sod all else

I just don't understand the systems and how they rate it all. Martins tap in was about the same chance as pen imo

I agree that the somewhat opaque and inconsistent methods used to calculate xG can make it hard to use. It's why I use four different sources to calculate an average, which would then be what I'd refer to on here.

By using that average I am trying to reduce the impact of inconsistency and bias in the various methodologies used. As I say, E361 is particularly stingy, but he does good stuff with the data he has so I continue to use him. 

As to how it is calculated, it varies, but generally you look at thousands of similar shots in similar games, and from those can craft an estimate of how many times out of 100 that type of shot results in a goal. Most places like infogol and e361 will have a page somewhere explaining their methods...although few go into extreme detail.

Martin's tap in was a great chance and infogol gave it 0.54 xG. However, the ball was delivered at reasonable pace by Semenyo, Martin and the ball are both moving away from the goal. It's easy but it's not dead certain. We can probably all think of times when that chance goes begging, or over the bar.

Whenever a team scores 3+ goals, they tend to outperform xG. If those 3 goals came from penalties, and they had no other chances, then their xG would be about 2.4, but they've scored 3 goals. Most other chances will get a lower rating than a penalty, and so you're always going to out perform your xG when you score heavily. Fulham put 6 past us and had xG of 2.13. 

Ultimately just remember that a) xG isn't perfect, it's not science, it's just better than simply looking at the number of shots taken and allowed, and b) it's best used to look at long term performance v results rather than single matches and who "should" have won.