DC - Postmatch

[Marito]

The team took 58 shots in 22 games, and according to the metric you value, only 2 of them had more than a minimal likelihood of success. The team used to be better at generating good chances than that. Other teams regularly do better than that. The team has traded quality for volume.

The number might add up to the same but it isn't exactly the same, is it?

Four walks are not the same as a home run and 3 strikeouts, even though they come out to the same number of total bases divided by plate appearances. Situationally, four walks might be better in some games while the home run is preferable in another. Always, a home run guarantees at least one run while walks never do, absent more contextual information. There is also substantial value in never making an out. But it does not wash out to exactly even despite the power percentage being exactly the same.

The same is true here. Contending that 10 .05 shots are the same as one 0.50 shot is silly, IMO. Further, it appears that teams are choosing to force us into this switch.

Bottom line, if we were dominating. If NYCFC were imposing its will on NER and DCU, then we would have seen both a lot of shots and a decent handful of high probability ones in each game. But that did not happen.

Maxi missed a 70% xG chance against DCU. So it's perfectly possible, even routine, to miss chances of that kind. A 50% xG chance means you are as likely to miss it as to score. Let's say you have two of those 50% or more xG chances in a game, and one comes from a defensive blooper that leaves your striker 1-on-1 without much effort on his part. And for the rest of the 90 minutes you have absolutely nothing that is quantifiable by xG, not even a super long distance shot that lands high up in the stands. Nada. No offensive production whatsoever. Would you prefer that than to have a team that shows some persistence and produces 10 or 12, 0.10 xG shots? Would you consider that more dominant?

I haven't seen the combined xG score for our DC game, but based on Doyle's twitter and knowing there were 31 shots, I'm guessing somewhere around 2 to 2.5. And DC's must have been 0.2. So the expected score, for all it's worth, was probably 2-0. We would all be happy now if expectations had been fulfilled.

 

[mgarbowski]

Maxi missed a 70% xG chance against DCU. So it's perfectly possible, even routine, to miss chances of that kind. A 50% xG chance means you are as likely to miss it as to score. Let's say you have two of those 50% or more xG chances in a game, and one comes from a defensive blooper that leaves your striker 1-on-1 without much effort on his part. And for the rest of the 90 minutes you have absolutely nothing that is quantifiable by xG, not even a super long distance shot that lands high up in the stands. Nada. No offensive production whatsoever. Would you prefer that than to have a team that shows some persistence and produces 10 or 12, 0.10 xG shots? Would you consider that more dominant?

I haven't seen the combined xG score for our DC game, but based on Doyle's twitter and knowing there were 31 shots, I'm guessing somewhere around 2 to 2.5. And DC's must have been 0.2. So the expected score, for all it's worth, was probably 2-0. We would all be happy now if expectations had been fulfilled.

I honestly don't see how this addresses the specific points I made at all. OTOH, it includes basic elements of a generic xG explanation that one would direct to someone who was unfamiliar with the concept, rather than someone who is making a very specific criticism of it.

Was I unclear? I don't think I was unclear.

ETA: I made 2 arguments:
1 - The idea that xG perfectly captures expected value and that you can add up separate combinations of xG shots interchangeably, as long as they all add up to the same figure, is at best a useful fiction.
2- The dichotomy between many low-value shots and fewer but higher value shots is false, and not necessary. If the team truly plays well, it will both create many shots, and a higher percentage than 2% will be high value. In my view, the team is failing to create high percentage shots and knows it, so it is compensating through volume, which I do not consider a sign of strength. It is merely better than doing nothing.

 

[Seth]

Also, that video reminded me about the national anthem singer. Will anyone disagree if I claim it was the worst, most off-key rendition we have heard in 4 years?

*stands up, shakes fist* I shall disagree mgarbowski !

Well, OK, yeah, it was pretty bad. The problem with the anthem as a piece of music is that it's hard to sing, period, let alone a cappella. Add in a super time delayed stadium echo and that can totally throw someone off. You're already on "by the dawn's early light" when "O say can you see" comes blasting at you through the stadium PA system. It's a nightmare to perform and honestly I don't see how anyone makes it through it.

Now to be fair to the many other anthem singers, most have done a lot better. It's not that she was off key, it's that she never really picked one. If you spend your life singing to music and are then presented with no music to follow but instead a time delayed echo of the previous line, well, things can get hairy. And for her, they did.

I'm not really a singer, but you couldn't pay me enough to do the anthem in a stadium. The sheer pressure to get the words right might alone do me in, let alone having to actually do a good job singing it.

All that being said, I don't think the anthem should precede sporting events. Might be a different discussion though.

 

[mgarbowski]

It's not that she was off key, it's that she never really picked one.

On this, we agree. She was all over the place.

 

[The Toe]

On this, we agree. She was all over the place.

We all thought she was worse than Bernie, and Bernie didn't even have to sing.

 

[SoupInNYC]

Maxi missed a 70% xG chance against DCU. So it's perfectly possible, even routine, to miss chances of that kind. A 50% xG chance means you are as likely to miss it as to score. Let's say you have two of those 50% or more xG chances in a game, and one comes from a defensive blooper that leaves your striker 1-on-1 without much effort on his part. And for the rest of the 90 minutes you have absolutely nothing that is quantifiable by xG, not even a super long distance shot that lands high up in the stands. Nada. No offensive production whatsoever. Would you prefer that than to have a team that shows some persistence and produces 10 or 12, 0.10 xG shots? Would you consider that more dominant?

I haven't seen the combined xG score for our DC game, but based on Doyle's twitter and knowing there were 31 shots, I'm guessing somewhere around 2 to 2.5. And DC's must have been 0.2. So the expected score, for all it's worth, was probably 2-0. We would all be happy now if expectations had been fulfilled.

I saw that 70% xG chance listed there. But if my memory serves me right, is that referring to the flying scissor kick that Maxi attempted? If so, yeah, that specific spot would have a high xG. But that specific play shouldn't. And I think that is one of the asterisks that is associated with the xG stat.

Or is that shot referring to another play?

 

[Seven]

I saw that 70% xG chance listed there. But if my memory serves me right, is that referring to the flying scissor kick that Maxi attempted? If so, yeah, that specific spot would have a high xG. But that specific play shouldn't. And I think that is one of the asterisks that is associated with the xG stat.

Or is that shot referring to another play?

No, that’s the scissor kick play. And yes, that seems like a lower percentage shot to me.

I was surprised how low the percentage was on the Callens header in the first half, from the Amagat cross. Seemed like a free header in a good spot, Callens just put it wide

 

[SoupInNYC]

No, that’s the scissor kick play. And yes, that seems like a lower percentage shot to me.

I was surprised how low the percentage was on the Callens header in the first half, from the Amagat cross. Seemed like a free header in a good spot, Callens just put it wide

They also had the Taty first half shot lower than I thought it would be. The one where he went near post and Hamid pushed it over.

 

[dummyrun]

They also had the Taty first half shot lower than I thought it would be. The one where he went near post and Hamid pushed it over.

For what it's worth, ASA's xG model liked the Callens and Taty shots more than Opta and the Maxi shot less. Opta does this weird thing where analysts subjectively designate some shots "big chances" and bump their xG, which is why you'll frequently see Baer's maps showing one or two huge circles and a bunch of tiny ones.

 

[canchon]

https://twitter.com/x/status/1039584922812465152

Hey look a team can even be worse finishing than NYCFC

 

[mgarbowski]

For what it's worth, ASA's xG model liked the Callens and Taty shots more than Opta and the Maxi shot less. Opta does this weird thing where analysts subjectively designate some shots "big chances" and bump their xG, which is why you'll frequently see Baer's maps showing one or two huge circles and a bunch of tiny ones.

Do they list individual shot data on their site or do you have non-public access?

 

[dummyrun]

Do they list individual shot data on their site or do you have non-public access?

They let me into the secret club, but the website has date filters that'll get you down to the player-by-game level. It's actually pretty amazing what you can do with ASA's public data.