SKC Postmatch

[canchon]

ETA: If SKC had higher expected goals than NYC in that game, it was only through the accumulation of massive numbers of really bad shot attempts - 18 shots, 3 on target.

I’m still trying to understand how the xG from SKC was so high. I remember 3 clear chances:
1. Espinoza’s header that went wide
2. Croizet’s run and touch through the middle (and that would have been ridiculous if it went in)
3. The free kick off the crossbar

I thought NYCFC had more strong opportunities, and not sure how they lost the xG score. Maybe it just is the huge amount of shots like Gator mentioned.

 

[SoupInNYC]

I’m still trying to understand how the xG from SKC was so high. I remember 3 clear chances:
1. Espinoza’s header that went wide
2. Croizet’s run and touch through the middle (and that would have been ridiculous if it went in)
3. The free kick off the crossbar

I thought NYCFC had more strong opportunities, and not sure how they lost the xG score. Maybe it just is the huge amount of shots like Gator mentioned.

Agreed, those are the only three real chances that I thought they had. And even though Croizet's was from such a dangerous spot, the amount of skill that was required to pull that off to put it on frame was super high.

I'm actually curious to see what the xG difference is from that shot, to the one that David had on the left side that Melia got to the ground on and made a hell of a one-hand save.

 

[Christopher Jee]

I’m still trying to understand how the xG from SKC was so high. I remember 3 clear chances:
1. Espinoza’s header that went wide
2. Croizet’s run and touch through the middle (and that would have been ridiculous if it went in)
3. The free kick off the crossbar

I thought NYCFC had more strong opportunities, and not sure how they lost the xG score. Maybe it just is the huge amount of shots like Gator mentioned.

Agreed, those are the only three real chances that I thought they had. And even though Croizet's was from such a dangerous spot, the amount of skill that was required to pull that off to put it on frame was super high.

I'm actually curious to see what the xG difference is from that shot, to the one that David had on the left side that Melia got to the ground on and made a hell of a one-hand save.

I kinda get why we are fixated on this but also not. If you are trying to look at the stars with reading glasses and you're not seeing them properly, it doesn't make sense to spend time trying to clean your reading glasses. And if you can't afford a telescope maybe you just sit and enjoy the view.

xG is a deeply flawed tool. It doesn't take into account player positions or the flight of the ball or a multitude of things that we know to be important factors in what makes a goal difficult to score.

Apparently club performance analysts are trying to get together to have one another's clubs install the equipment necessary to measure the things we need to measure to come up with better tools. Even then, the amount of manual labor that goes into classifying the raw data is absurd, and even top clubs don't necessarily have performance analysts. I would expect this dynamic to be disrupted in the near future with the cost of equipment going down and the availability of computer vision technology to take a lot of the grunt work out of classification.

Until then I'll just marvel at the night sky

 

[sammy3469]

Obviously, the Audi index does not measure being outrun by Ben Sweat by 20 yards in a blind sprint.

ETA: If SKC had higher expected goals than NYC in that game, it was only through the accumulation of massive numbers of really bad shot attempts - 18 shots, 3 on target.

FWIW, Whoscored had NYCFC with a higher rating than SKC (7 v 6.62), so there's at least one ratings system that disagrees with Audi and xGo. It also really liked Herrera's match (4 aerial duels won, 6 tackles, 4 interceptions)...of course he also gave the ball away 6 times.

 

[Gotham Gator]

Expected goals looks at the position of a shot and runs an analysis of how often other shots from that spot score. There may be other things involved, but I do not believe the measure takes into account where the defenders are positioned (could be wrong there).

SKC had a ton of shots blocked. Some of those were from pretty good spots in the box, but we had a lot of defenders on the scene. These were not white knuckle moments for us as fans, because we could clearly see that the shooter was defended well. However, once the shot goes off, it counts toward SKC's expected goals total.

Think about this as well. David Villa had that wide open break at the end of the half. The ball was taken off his foot just as he was shooting. That was more dangerous than almost any chance SKC had, but it counted zero toward our expected goals total.

I like expected goals as a stat, especially over time, but it has its limits.

 

[mgarbowski]

Surprised this hasn't been posted yet. Behind the Scenes Video from SKC.

 

[Midas Mulligan]

I do not believe the measure takes into account where the defenders are positioned (could be wrong there).

I believe it does.

Location, defenders between goal, how the ball arrives (ie, chance on foot or head) are all factors.

It is graded very manually by Opta, though. There’s some good explanations re Opta technique.

 

[Davo15]

I'm an old Fart. I trust my eyes or match reports. Have never looked at any of this Audi or opta stuff don't know how any of it works and flibbity flabbity flu that's the way I like it.

 

[SoupInNYC]

I don't think this has been brought up yet, but the use of VAR in this game was EXACTLY how it should be done. Quick, decisive, and getting calls correct. Fantastic job by Geiger on this.

 

[dummyrun]

Just a general reminder that there are a bunch of different xG models out there that have different methodologies. For example:

https://cartilagefreecaptain.sbnati...ier-league-projections-and-new-expected-goals

https://thestatattackblog.wordpress.com/2014/07/09/updated-expected-goals/

http://11tegen11.net/2015/08/14/a-close-look-at-my-new-expected-goals-model/

https://differentgame.wordpress.com/2017/04/29/an-xg-model-for-everyone-in-20-minutes-ish/

 

[dummyrun]

I don't think this has been brought up yet, but the use of VAR in this game was EXACTLY how it should be done. Quick, decisive, and getting calls correct. Fantastic job by Geiger on this.

I bet this has been brought up but VAR is actually always extremely bad

 

[Seven]

https://twitter.com/x/status/971436530421055489

Maybe further explains him being "off"

Is a "hard" cold just a language thing?

 

[EganSoccerWords]

edit: nm just up for the saturday games

 

[Gotham Gator]

https://twitter.com/x/status/971436530421055489

Maybe further explains him being "off"

Is a "hard" cold just a language thing?

Ha! Maybe.

Rojas is one of the smartest, most astute analysts covering the team, but sometimes that gets lost in translation.

 

[canchon]

Ha! Maybe.

Rojas is one of the smartest, most astute analysts covering the team, but sometimes that gets lost in translation.

He’s much more insightful then some of those other blowhards who can’t remember what player did what or players names on SITC.

His podcast in Spanish is great too - w/o some of the translation issues.

 

[Gotham Gator]

On the subject, I am now working through the latest SITC. One interesting tidbit is that Coach V started Sweat for the simple reason that he is ahead on the depth chart right now based on pre-season performance.

 

[joe]

His podcast in Spanish is great too - w/o some of the translation issues.

What's it called?

 

[LionNYC]

On the subject, I am now working through the latest SITC. One interesting tidbit is that Coach V started Sweat for the simple reason that he is ahead on the depth chart right now based on pre-season performance.

Every player we've signed to an extension and underperformed in that following year. Mainly Saunders. And also McNamara comes to mind. And RJ Allen.

That being said, I hope it doesn't happen to Villa and Matarrita who both signed extensions for this season.

 

[canchon]

What's it called?

Jugador Franquicia

The one annoying thing is that they play way too much intro music at the beginning. But it’s a Spanish-language mls podcast which is great.

 

[Kjbert]

What's it called?

Soccero en la citio