Page 22 of 41

Re: stat attack!

PostPosted: Mon Jul 27, 2020 12:54 pm
by Sideshow Bob
leroy was faster. :(

Re: stat attack!

PostPosted: Mon Jul 27, 2020 2:54 pm
by patrickblue
Sideshow Bob wrote:leroy was faster. :(


Don't think he was. And I am/was Leroy's biggest fan.

Re: stat attack!

PostPosted: Mon Jul 27, 2020 6:27 pm
by Sideshow Bob
patrickblue wrote:
Sideshow Bob wrote:leroy was faster. :(


Don't think he was. And I am/was Leroy's biggest fan.


kdb thought leroy was faster (@ 9:32): https://www.youtube.com/watch?v=dRp4DpWJb3c

side note: i'd wager alphonso davies would smash them both. he is ridiculous.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 12:02 am
by PrezIke
Saul Goodman wrote:Our record this season with Gundogan starting is 53% while its 80% when he doesnt start.
There are of course other things (like opponent) to factor in but its an interesting stat


Sorry Saul. I mean no disrespect, but you know this is a mostly meaningless stat out of context that's red meat to all his detractors.

There are many other variables to consider....HOWEVER....

One cannot deny Gundo did not have his best season and we need better from him.

WhoScored.com says he had a 6.70 rating for the PL in 20 starts and 10 off the bench. That's ranked 17th for the entire team. That is below Angelino, Otamendi and Mendy.

That's poor for a potential starting player on a top club. He had 1 assist and 2 goals. Last season he was given MOTM 2 times. This? None. He fared far better in CL with their average of 7.33 in 7 appearances.

https://www.whoscored.com/Players/77464 ... 3%BCndogan

https://understat.com/player/314

His xA (Expected assists) per 90 minutes dropped from last season to .13 from .21 despite his xA being better than the actual output.

However, it's worthy to note that what some of us who defend Gundo constantly bring up to defend him is also supported by the stats.

The graph shows how his skill in xBuildup per 90 min "build up play" and xChain per 90min (which is xG that came from possessions he was involved with) is evident.

His xBuildup is .68 per 90 compared to the average of .23 per 90.
His xChain per 90 was .86 to the average of .34

He actually improved slightly compared to 2018-19 in both xBuildup and xChain.

Even though his assists dropped his xBuildup stats remain amongst the best in the league.

The contention of some who defend him is that build up play comes from his ball retention skills but we don't often see so clearly and is underrated by the average eye test compared to those who contend it as vital to success.

Yaya was a master of it along with Silva and was why we had so many attacking situations.


Now both are gone. Gundo is up there along with Foden and Bernardo who both have a ridiculous first touch, but more more rapidly and don't slow the play down as much when needed.

Oddly enough Silva's lack of respect for much of his City career was similarly maligned by the reasons we slate Gundo. Silva's turns and other attributes have made him loved, but we may need Gundo next season as a similar protector of the ball.

As some have said, playing Gundo with Rodri and I also contend Silva, plus our frantic defenders led all of their weaknesses to become more glaring than they should.

All of those three lack pace that clearly caused us all sorts of problems against a lot of opposition when we are already a side vulnerable to counter attacks given how high of a line we have and intent to attack.

He'll play more at the 6 next year and we will have more pace and stability in the back line. With that I suspect you will find his skills more useful.

Remember, his ball retention skills is one of the reasons he almost always plays against teams that press high.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 12:37 am
by PrezIke
Some interesting stats:

https://understat.com/league/EPL/2019

Expected points in the PL 2019-20

City: 86.76
Liverpool 74.38


Think we don't take our chances?

City xG in the PL: 102.21
City Goals: 102

Re: stat attack!

PostPosted: Tue Jul 28, 2020 12:51 am
by PrezIke
Based on WhoScored's best player average ratings for the season:

https://www.whoscored.com/Regions/252/T ... -2019-2020

1) Kev - 7.97
2) Ricardo Pereira - 7.50
3) Adama Traoré - 7.49
4) Mahrez - 7.48
5) Mane - 7.45
6) Salah - 7.40
7) Raz - 7.35
8) Trashford - 7.35
9) Maddison - 7.34
10) Kane 7.33
.
.
.
.
.
32) Jordan Henderson - 7.15

Re: stat attack!

PostPosted: Tue Jul 28, 2020 12:59 am
by PrezIke

Re: stat attack!

PostPosted: Tue Jul 28, 2020 6:02 am
by BlueinBosnia
PrezIke wrote:Think we don't take our chances?

City xG in the PL: 102.21
City Goals: 102

I'd assume that factors in the proportion of our chances that are converted to goals. So basically it's not saying that we're taking our chances, but rather our finishing wasn't any better or worse than expected, statistically speaking.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 7:16 am
by Bluemoon4610
PrezIke wrote:Some interesting stats:

Think we don't take our chances?

City xG in the PL: 102.21
City Goals: 102

Shouldn't this be posted on the finishing thread? :?:

Re: stat attack!

PostPosted: Tue Jul 28, 2020 9:46 am
by dazby
It doesn't fit with Nigel's narrative, so no.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 9:56 am
by Mase
What does City xG mean?

Re: stat attack!

PostPosted: Tue Jul 28, 2020 10:28 am
by Foreverinbluedreams
Mase wrote:What does City xG mean?


Expected goals.

All chances/shots are marked out of 1. A worldy from 30 yards would have a 0.1 xg and an open goal tap in would have a 1 xg.

So being good at smashing them in from long range can hide the fact that you're really poor at finishing sitters.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 10:53 am
by Mase
Foreverinbluedreams wrote:
Mase wrote:What does City xG mean?


Expected goals.

All chances/shots are marked out of 1. A worldy from 30 yards would have a 0.1 xg and an open goal tap in would have a 1 xg.

So being good at smashing them in from long range can hide the fact that you're really poor at finishing sitters.


Thank you

Re: stat attack!

PostPosted: Tue Jul 28, 2020 2:08 pm
by PrezIke
Mase:

This website has been keeping these stats and is fantastic to get another perspective on player and team performances:

https://understat.com/

Expected goals (xG) is the new revolutionary football metric, which allows you to evaluate team and player performance.

In a low-scoring game such as football, final match score does not provide a clear picture of performance.

This is why more and more sports analytics turn to the advanced models like xG, which is a statistical measure of the quality of chances created and conceded.

Our goal was to create the most precise method for shot quality evaluation.

For this case, we trained neural network prediction algorithms with the large dataset (>100,000 shots, over 10 parameters for each).

On this site, you will find our detailed xG statistics for the top European leagues.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 2:12 pm
by PrezIke
dazby wrote:It doesn't fit with Nigel's narrative, so no.


TBF, the problem with both total xG and total goals scored is that its a TOTAL.

I have to go look, but we all know we had several games where we scored 4+ goals.

The issue is there are some games where where we struggle to score.

My observation is two other factors causing us problems (which likely signal something else as well):

1) When we don't score early we more frequently struggle to score at all (this may come from a variety of things)

2) If we fall behind in the 1st half we also appear to struggle to score.

Re: stat attack!

PostPosted: Tue Jul 28, 2020 2:16 pm
by Mase
PrezIke wrote:Mase:

This website has been keeping these stats and is fantastic to get another perspective on player and team performances:

https://understat.com/

Expected goals (xG) is the new revolutionary football metric, which allows you to evaluate team and player performance.

In a low-scoring game such as football, final match score does not provide a clear picture of performance.

This is why more and more sports analytics turn to the advanced models like xG, which is a statistical measure of the quality of chances created and conceded.

Our goal was to create the most precise method for shot quality evaluation.

For this case, we trained neural network prediction algorithms with the large dataset (>100,000 shots, over 10 parameters for each).

On this site, you will find our detailed xG statistics for the top European leagues.


Nice one

Re: stat attack!

PostPosted: Fri Aug 07, 2020 9:05 pm
by PrezIke

Re: stat attack!

PostPosted: Fri Aug 07, 2020 9:38 pm
by Foreverinbluedreams

Re: stat attack!

PostPosted: Mon Sep 21, 2020 9:09 pm
by nottsblue
Another record

First side ever to win 10 opening day games in a row

Re: stat attack!

PostPosted: Mon Sep 21, 2020 9:15 pm
by PrezIke