Big thread on Championship stats and analysis...

[spudski]

....some really interesting analysis on this twitter thread. 

The one that struck me re City was our ineffectiveness at crossing compared to the rest of the league. 

Effective cross being defined by a shot or key pass after 5 seconds of the cross. 

 

[Rob k]

Only had a very brief look - however my initial thought is, as clubs look heavily at stats, will we get the 25m + we want for Scott? 
 

[spudski]

15 minutes ago, Rob k said:

Only had a very brief look - however my initial thought is, as clubs look heavily at stats, will we get the 25m + we want for Scott? 
 

True...but he obviously has immense talent, still young, and can continue improving. 

He does well on many stats. 

Topped most fouls drawn against a player in the Championship. 

[Rob k]

9 minutes ago, spudski said:

True...but he obviously has immense talent, still young, and can continue improving. 

He does well on many stats. 

Topped most fouls drawn against a player in the Championship. 

And he's under contract to be fair which helps 

Edited May 12, 2023 by Rob k

[downendcity]

Ineffectiveness at crossing has been in our DNA for years.

Pring and Williams trying to reverse the trend at QPR though!

[spudski]

44 minutes ago, downendcity said:

Ineffectiveness at crossing has been in our DNA for years.

Pring and Williams trying to reverse the trend at QPR though!

What's interesting about the stats are...

'Over 30% of all open-play goals in Championship come from crosses (directly or indirectly), making them an important route to scoring. Here's how teams are ranked by overall crossing effectiveness.'

What those stats don't show, is how many attempted crosses are made in a game. Stats have shown over the years that it takes over 90 crosses to achieve a goal. So although 30% of goals scored have come from crosses...it's still not an affective way of scoring percentage wise, as it takes on average approx 90 crosses to score a goal. 

If you look at the creation analysis it's interesting, as it shows ours is predominantly from wide areas. Note how hardly any team creates from directly Infront of the 18yard 'D'. It's very low. I've always seen this in previous years, and thought it was their to be exploited...playing and creating centrally. 

 

 

[ExiledAjax]

15 minutes ago, spudski said:

What's interesting about the stats are...

'Over 30% of all open-play goals in Championship come from crosses (directly or indirectly), making them an important route to scoring. Here's how teams are ranked by overall crossing effectiveness.'

What those stats don't show, is how many attempted crosses are made in a game. Stats have shown over the years that it takes over 90 crosses to achieve a goal. So although 30% of goals scored have come from crosses...it's still not an affective way of scoring percentage wise, as it takes on average approx 90 crosses to score a goal. 

If you look at the creation analysis it's interesting, as it shows ours is predominantly from wide areas. Note how hardly any team creates from directly Infront of the 18yard 'D'. It's very low. I've always seen this in previous years, and thought it was their to be exploited...playing and creating centrally. 

 

 

Yeh it's a repeat of the Reep fallacy of looking at the stats in reverse. Just because goals are scored in that way, doesn't mean it's a good way to score goals.

You also need to look at what happens to the crosses that don't become goals. Most will either become goal kicks (conceding possession), be punched or headed into open play (contested possession) or become corners (retain possession but corners have an even lower goal conversion rate than open play crosses).

In almost every case the outcome for a cross that does not become a goal is a negative one for the attacking team.

Personally the "best" way to score goals is to find the highest goal return from the lowest effort and lowest risk attacking play. That may well end up being crosses, but I'd be surprised.

[spudski]

28 minutes ago, ExiledAjax said:

Yeh it's a repeat of the Reep fallacy of looking at the stats in reverse. Just because goals are scored in that way, doesn't mean it's a good way to score goals.

You also need to look at what happens to the crosses that don't become goals. Most will either become goal kicks (conceding possession), be punched or headed into open play (contested possession) or become corners (retain possession but corners have an even lower goal conversion rate than open play crosses).

In almost every case the outcome for a cross that does not become a goal is a negative one for the attacking team.

Personally the "best" way to score goals is to find the highest goal return from the lowest effort and lowest risk attacking play. That may well end up being crosses, but I'd be surprised.

It's an interesting thought re your last paragraph. Different types of players excel in converting different types of goal-scoring chances. For example, some players excel in beating the defensive line and dribbling towards goal, whereas other players prefer high crosses into the box. Thus, players who are provided with types of goal-scoring chances they do not favor might score fewer goals than their expected goals suggests. For instance, if Lionel Messi were provided were provided with high crosses instead of low passes, he would likely score less than expected.

From the stats re us...is it the poor quality of crosses, or that the players up front are better with a pass to feet, rather than a cross. 

Would we be more effective going through the middle, with the types of players we have? 

[ExiledAjax]

4 minutes ago, spudski said:

Would we be more effective going through the middle, with the types of players we have? 

We were at our best 'this' season going through the channels. Pring, Semenyo, Conway, those powerful runners moving quickly through the gap between the full back and the centre back , down the edge of the penalty box, or just to either side of it. Down to the bye line and then a low cross/cut back towards the space between the penalty spot and the 6-yard box. Not sure if the stats show that but when we were threatening, that was an effective tool for us. 

We've lost Semenyo, and he was excellent at this, but personally that is a hard to defend move that has a high probability of a goal and if it's not a goal then it's not guaranteed to cede possession.

[IAmNick]

What is a cross?

I think for it to be really useful you'd have to break it down by position... clearly something ballooned in from 30 yards out is not the same as a cut back from the byline in the penalty area. It doesn't really make sense to group them as chance types to me.

Part of our problem imo has been the poor quality of crosses. Pring has been a breath of fresh air using his strength and determination to get into the box and create high value crosses, as opposed to a no look smash from miles away because you're not sure what to do next.

Edited May 12, 2023 by IAmNick

[spudski]

16 minutes ago, ExiledAjax said:

We were at our best 'this' season going through the channels. Pring, Semenyo, Conway, those powerful runners moving quickly through the gap between the full back and the centre back , down the edge of the penalty box, or just to either side of it. Down to the bye line and then a low cross/cut back towards the space between the penalty spot and the 6-yard box. Not sure if the stats show that but when we were threatening, that was an effective tool for us. 

We've lost Semenyo, and he was excellent at this, but personally that is a hard to defend move that has a high probability of a goal and if it's not a goal then it's not guaranteed to cede possession.

I agree with that. 

I look at Bell, Conway, Wells and Weimann...even Sykes...imo they would be suited more to this approach, rather than crosses from wide. 

Yes we've scored doing it... however we maybe more effective going more centrally. 

[Davefevs]

On a slight tangent, have a watch of this:

And a read of this:

This really is starting to get into solving the gap between Eyes and Data, but giving “Eyes at Scale” (we can’t watch every match, every player, every run, etc…and better still, turning it into data.

Statsbomb are doing something similar to this with Statsbomb 360.

 

This stuff is where you really get to see what a Matty James or an Andi Weimann brings.  That’s not a criticism of fans, it just not all fans watch the game in the same way, they also have different reasons for watching too.

But if you are more of the analytical persuasion, this would be gold-dust.

 

[downendcity]

1 hour ago, spudski said:

What's interesting about the stats are...

'Over 30% of all open-play goals in Championship come from crosses (directly or indirectly), making them an important route to scoring. Here's how teams are ranked by overall crossing effectiveness.'

What those stats don't show, is how many attempted crosses are made in a game. Stats have shown over the years that it takes over 90 crosses to achieve a goal. So although 30% of goals scored have come from crosses...it's still not an affective way of scoring percentage wise, as it takes on average approx 90 crosses to score a goal. 

If you look at the creation analysis it's interesting, as it shows ours is predominantly from wide areas. Note how hardly any team creates from directly Infront of the 18yard 'D'. It's very low. I've always seen this in previous years, and thought it was their to be exploited...playing and creating centrally. 

 

 

It is highly likely that the reason that it takes 90 crosses to produce one goal is that most teams are pretty poor at crossing. Even at premier league level 9 elite level players) I never cease to be amazed at how many crosses/freekicks/corners fail to clear the first man.

I remember when most teams played with "proper" wingers, whose job it was to get to the byline and get crosses in. Back  then my recollection is ( might be selective memory though!) that a far greater percentage of crosses got into the danger area, although I doubt there are any statistics to show whether the ratio of crosses:goals was better or worse than today's.

Of course, back in the day tactics were a lot different. Teams would set up with a big target man type striker, who was good in the air (think John Galley) and would play players capable of getting to the byline and getting crosses into the box for the big striker to get on the end of. Pep would have kittens at the thought of those tactics!

Even when modern players get to the byline it seems there is an increasing trend to drive the ball low into the goal area rather than cross in a way that encourages a header.

[Davefevs]

@downendcity I don’t have the stats either, but I reckon going back to the likes of Steve Coppell and Peter Barnes that more crosses were chipped rather than whipped.

Even looking at corners, the Gary McAllister / Sami Hyypia near post ball was a gently flighted cross, unlike the whipped variety to which, as you say, rarely beats the first man.

Different levels of precision / difficulty to execute I guess.  Different methods of defending them too I guess?

The Mark Little / Aaron Wilbraham opening day goal v Sheffield Utd was proper “old-skool” wasn’t it.

[spudski]

3 hours ago, downendcity said:

It is highly likely that the reason that it takes 90 crosses to produce one goal is that most teams are pretty poor at crossing. Even at premier league level 9 elite level players) I never cease to be amazed at how many crosses/freekicks/corners fail to clear the first man.

I remember when most teams played with "proper" wingers, whose job it was to get to the byline and get crosses in. Back  then my recollection is ( might be selective memory though!) that a far greater percentage of crosses got into the danger area, although I doubt there are any statistics to show whether the ratio of crosses:goals was better or worse than today's.

Of course, back in the day tactics were a lot different. Teams would set up with a big target man type striker, who was good in the air (think John Galley) and would play players capable of getting to the byline and getting crosses into the box for the big striker to get on the end of. Pep would have kittens at the thought of those tactics!

Even when modern players get to the byline it seems there is an increasing trend to drive the ball low into the goal area rather than cross in a way that encourages a header.

 

50 minutes ago, Davefevs said:

@downendcity I don’t have the stats either, but I reckon going back to the likes of Steve Coppell and Peter Barnes that more crosses were chipped rather than whipped.

Even looking at corners, the Gary McAllister / Sami Hyypia near post ball was a gently flighted cross, unlike the whipped variety to which, as you say, rarely beats the first man.

Different levels of precision / difficulty to execute I guess.  Different methods of defending them too I guess?

The Mark Little / Aaron Wilbraham opening day goal v Sheffield Utd was proper “old-skool” wasn’t it.

I've read that it's the defence that's been coached and better disciplined that negates crossing. 

Watch how much space they had in the past.