Stats and Analysis Thread

[darlowolf64]

It isn't great to concede first but it's the result at the END of the match that matters, surely?

 

[Deleted Cyber]

Playing ok and creating fuck all still means you don’t win .

6 points from Leeds when stats said two draws was probably right and both goals were flukes.
6 points from Fulham , who missed a sitter and we scored in added time each team had three shots on target in the second game Fulham has 14 to our 7 overall. Dodgy VAR decision aside could not have grumbled at 0 0 in both so that’s 8 points off already .
Sheffield at home was very low on quality and the goal we scored was great .
I guess the six games you think we should have got something would be Liverpool Lecister Man United Aston Villa where we played ok and lost 1 0 ( again if you don’t score you can’t win however we’ll you play) the other two? Everton possibly a draw at best but Newcastle a draw in each was right as they were better once and we were better once? We deserved fuck all from West Ham Burnley West Brom Man City , got what we deserved re saints ,spurs palace both games, arsenal , Chelsea, Brighton we ended up holding on for a point. So we are probably 5 or 6 points better off than we should be with our performances and statistics. We have been generally awful in the final third, mediocre in midfield and generally ok at the back . Defensively when we fuck it up we go the whole hog and do it big style! Offensively we have score three once all season and included a penalty and og in those!
it’s alright. I will go and give my daftie head a wobble or whatever today’s dismissive phrase is.

 

[Throbber3]

It isn't great to concede first but it's the result at the END of the match that matters, surely?

Surely statistics show that in the majority of premier league games the team who scores first is more likely to win? This is a little out of date but shows that the first goal is very important in the premier league.

 

[Deutsch Wolf]

I honestly haven't got a clue what you're on about, mate.

We've won 11 games, each of the bottom three have won 5. We're objectively miles better than them, even while underperforming (no-one is going to claim we've played well).

Can I ask what this 21 out of 23 stat is please? Because we've kept six clean sheets in 2021 alone.

 

[AndyWolves]

Really?
Worst expected goals per shot 0. 08. Rare as hens teeth showing we are not creating anything dangerous ie powderpuff.
We score great goals or nothing it seems .
We are below the prolific Sheffield United. ( ps they’re bottom and relegated)
Passes into the area we are also bottom. Again creating no danger so creating no chances. You can’t blame WJ if we aren’t giving him chances. ( below WBA who are second bottom)
Our pressure in the attacking third we are bottom so we sit too deep according to the stats and don’t press high enough.
Low on ball recovery. 2nd and third balls won by the opponents. Blocked passes very low if not bottom which is fine if we are a high possession side but we are not. This was our strength in previous seasons and it’s really dropped this season.
9 of 11 wins by 1 0
Scored first twice in 23 games.
Looks more like relegation form to me, not silly. We are fortunate playing how we have this year to be safe with 5 to go

Last season that figure was 0.10 which is probably the difference having Raul would make, or maybe even some of Silva's shots going in.

Our pressures last year in the attacking 3rd were second bottom behind Newcastle.

The pressures thing is a bit mis-leading, we apply less pressure than most other teams but our successful pressures are roughly the same which means we're being more 'efficient'. That said, our pressures and recoveries are both down against last season.

I need to re-watch this video again and compare to last season to get a better idea of how this all stacks up and how much better / worse we've been this season.

Shot creating actions we're 9th overall but goal creating actions we're 4th from bottom - last year we finished 10th for SCA and 9th for goal creating actions. Our shot creating actions are marginally higher this season. Our xG will be down this season too. Last season we were 1.24 xG per game and this season we're 1.05.

Our xA last season was 0.85 and this season it's 0.70. So we're creating less this season and last season we scored more goals where there wasn't an assist.

The other thing around pressures and recoveries, my own view on this is that Jonny, Jota and Jimenez were all excellent at applying pressure and winning the ball back. Without these guys in the team we press less and thus make fewer recoveries.

Goals conceded is interesting too, last season Rui saved 1.2 goals more than would be expected, this season he's conceded 4.3 more than expected.

My collective view on this data is that we're suffering from not scoring as many goals as last season (and not expecting to score as many goals). This could probably be explained, if a little simplistically, by not having Raul J. WJ's xG is the same as Podence / Neto - as our frontman you'd expect it to be higher. We've conceded more goals than we should have and we're also pressing less with less recoveries.

I think having Raul back would make a tremendous difference to our side and would mitigate some of the other weaknesses in the team.

I also think it shows what fine margins are between an average season and a relatively poor one. If we'd scored all out "expected goals" we'd be on 5 more goals, and if we hadn't conceded more than "average" which we're over by 4. Last season we over-achieved on both counts. This would represent a "swing" of 9 goals (or an Emi Martinez contribution for Villa) and surely a few more points in our favour and having us level or nearer Leeds, Arsenal and Villa.

 

[YoungWolf]

Wow
The stats completely back up how many areas we need to improve on to survive. With those stats we should probably have gone down . I can’t see us being that fortunate again so hopefully lessons learned

Not even close

 

[AndyWolves]

In defence of Adama Traore.

Adama has had a really poor season from a goals and assists perspective but those two stats don't tell the whole story.

What the stats do show is in regards to shot creating actions (pass, dead ball, dribble, shot, being fouled or a defensive action) which contribute to a team mate having a shot. So an assist to the assister kind of thing.

This is where Adama tops the charts, he has the most shot creating actions (SCA) when free kicks are removed. (Incidentally, this metric shows how ace Pedro Neto is). So whilst we don't see this output in terms of goals and assists directly attributable to Adama, what we do see is his influence in allowing wolves to get into a situation where a shot can occur.

Incidentally, his SCA this season is higher than it was last season. His goal contribution per game last season was 0.52, or, in every 1/2 games Adama did something which lead to a goal, however, this season it's 0.19, or 1/5 games.

In terms of progressive carries up the pitch, he comfortably leads that stat too (as you would expect!).

So, from this, I would surmise that Adama has a role in the team which is getting us to a position where we're able to have a shot on goal.

I don't really think I've noticed this too much when watching games (either this season or last season) so I find this quite an interesting perspective.

 

[darlowolf64]

In defence of Adama Traore.

Adama has had a really poor season from a goals and assists perspective but those two stats don't tell the whole story.

What the stats do show is in regards to shot creating actions (pass, dead ball, dribble, shot, being fouled or a defensive action) which contribute to a team mate having a shot. So an assist to the assister kind of thing.

This is where Adama tops the charts, he has the most shot creating actions (SCA) when free kicks are removed. (Incidentally, this metric shows how ace Pedro Neto is). So whilst we don't see this output in terms of goals and assists directly attributable to Adama, what we do see is his influence in allowing wolves to get into a situation where a shot can occur.

Incidentally, his SCA this season is higher than it was last season. His goal contribution per game last season was 0.52, or, in every 1/2 games Adama did something which lead to a goal, however, this season it's 0.19, or 1/5 games.

In terms of progressive carries up the pitch, he comfortably leads that stat too (as you would expect!).

So, from this, I would surmise that Adama has a role in the team which is getting us to a position where we're able to have a shot on goal.

I don't really think I've noticed this too much when watching games (either this season or last season) so I find this quite an interesting perspective.

Isn't 1 in 5 games worse than 1 in 2? So hes been over twice as shit this season unless you are including him providing a goal or SCA for the opposition...

 

[AndyWolves]

Isn't 1 in 5 games worse than 1 in 2? So hes been over twice as shit this season unless you are including him providing a goal or SCA for the opposition...

Depends whether you think he's responsible for his colleagues not scoring.

Adamas expected assists are 4.7, second highest at the club. In terms of actual assists compared to xA Adama is rock bottom on -2.7

The whole club are underperforming both against xG and xA, last season we over performed against both metrics.

 

[YoungWolf]

Depends whether you think he's responsible for his colleagues not scoring.

Adamas expected assists are 4.7, second highest at the club. In terms of actual assists compared to xA Adama is rock bottom on -2.7

The whole club are underperforming both against xG and xA, last season we over performed against both metrics.

We underperformed as a team xG for and against last season. As I have said, with slightly more luck we’d have been top six maybe five

 

[Johnny75]

I'm not sure xA is useful. If he's providing chances for Donk or Semedo for example then that player is more likely to miss.

If he's providing a dinked cross that too is unlikely to provide a good chance and I'm not sure the metric allows for that (it allows for crosses as a whole rather than type).

xA coupled with who the chance was for and type of cross/ pass would be a better metric to use but then your sample size would be tiny for comparison.

Tl;dr the xA or SCA stat is useless without context.

 

[AndyWolves]

We underperformed as a team xG for and against last season. As I have said, with slightly more luck we’d have been top six maybe five

Fbref shows us scoring 49 goals with an xG of 47.1

Have I missed something or misunderstood something?

 

[AndyWolves]

I'm not sure xA is useful. If he's providing chances for Donk or Semedo for example then that player is more likely to miss.

If he's providing a dinked cross that too is unlikely to provide a good chance and I'm not sure the metric allows for that (it allows for crosses as a whole rather than type).

xA coupled with who the chance was for and type of cross/ pass would be a better metric to use but then your sample size would be tiny for comparison.

Tl;dr the xA or SCA stat is useless without context.

That's the thing though, it seperates the player providing the assist from the quality of the shooter.

Otherwise you'd end up in a situation whereby players like Adama get criticism because they aren't providing assists to Raul (i.e. players who perform better against xG).

Obviously, there is nuance if Adama has the choice between assisting different players in the box and there aren't any stats to show that which is where observations come in.

 

[Johnny75]

That's the thing though, it seperates the player providing the assist from the quality of the shooter.

Otherwise you'd end up in a situation whereby players like Adama get criticism because they aren't providing assists to Raul (i.e. players who perform better against xG).

Obviously, there is nuance if Adama has the choice between assisting different players in the box and there aren't any stats to show that which is where observations come in.

You're proving my point here Andy.

 

[AndyWolves]

You're proving my point here Andy.

I don't follow, for me, using xA is a "fairer" metric than using actual assists. The reason being it's not Adamas fault that Donk heads over the bar, it's also not to Adamas credit if Raul scores an impossibility unlikely chance.

Adama doesn't really get to decide who the assist is for.

 

[Johnny75]

Providing assists is a useless metric unless it's coupled with who the assist is for.

It's also useless if the quality of the assist is crap delivery for the player.

I'm sure Adama will have been given a decent xA score for a chipped cross into Podence but there is no chance Pod scores as he is right now.

It renders the stat pointless.

 

[Tony Towner]

Do you know what's the best metric? Your eyes.

 

[Johnny75]

Do you know what's the best metric? Your eyes.

That's not true either.

It's a combination of both. Neither should be dismissed or held up as the one thing that works.

 

[AndyWolves]

Providing assists is a useless metric unless it's coupled with who the assist is for.

It's also useless if the quality of the assist is crap delivery for the player.

I'm sure Adama will have been given a decent xA score for a chipped cross into Podence but there is no chance Pod scores as he is right now.

It renders the start pointless.

It's about collecting data that can be used for comparison.

If we evaluated assists based on who's on the end of them then Dennis Pierce would look amazing. It's irrelevant when comparing which players generate more assists.

Saying Adamas xA are worth less because Podence is on the end of them misses the point. Likewise, an "inferior" assister would be scored higher for assisting Raul.

 

[AndyWolves]

Do you know what's the best metric? Your eyes.

My eyes, like everyone else's, are biased. This isn't about closing your eyes and relying purely on the data. It's about looking at the data to see if what your eyes tell you is true, or if you've missed something, or is there something else interesting happening?