The best league in the world?

SlowDownDerek said:
Lol, you are a massive hypocrite. I post lots of opinions alongside my posts that disagree with people, you can see someone responding to one of my opinions two posts above. You on the other hand are constantly guilty of what you accuse me of. Such as this post and all of the others that say the same thing. Also the one above which shows you didn't read DJs post properly in your haste to have a go.

Hypocrisy at it's worst.
Click to expand...
Lol.

I like to get involved in forum debate.

You seem to just like a fight for the sake of it.

Let's agree to disagree at this point, otherwise you'll continue with the gaslighting.
 
dragonm8 said:
Lol.

I like to get involved in forum debate.

You seem to just like a fight for the sake of it.

Let's agree to disagree at this point, otherwise you'll continue with the gaslighting.
Click to expand...

There is a debate going on in this thread, that I'm involved in, and you are here trying to start pointless arguments. People in glass houses.
 
Kudos said:
I don't think we're drastically underperforming against our xG it's more that we've been spoilt by fantastic goals raising the threshold for what we think a good chance looks like.
Click to expand...

Perhaps, probably also to do with previous styles of play that put such a premium on chances and keeping it tight. Including the last couple of seasons of Howe's reign.
 
SlowDownDerek said:
There is a debate going on in this thread, that I'm involved in, and you are here trying to start pointless arguments. People in glass houses.
Click to expand...
First orgasm you've had all season.

A debate, and an argument you're involved in at the same time.

The pinnacle.
 
Kudos said:
I don't think we're drastically underperforming against our xG it's more that we've been spoilt by fantastic goals raising the threshold for what we think a good chance looks like.
Click to expand...

I had a look as I didn't know. Using Understat data, we are actually second worse when comparing xG against goals scored. We have scored 52 but were expected to score 60.68. Everton seem to be doing very badly.

8 teams are overperforming (in blue), 12 underperforming (in white)

Capture.png

I'll look at defensive record and post this too
 
Matt Stevenson said:
I had a look as I didn't know. Using Understat data, we are actually second worse when comparing xG against goals scored. We have scored 52 but were expected to score 60.68. Everton seem to be doing very badly.

8 teams are overperforming (in blue), 12 underperforming (in white)

View attachment 14170

I'll look at defensive record and post this too
Click to expand...

With this though, we take a lot of low chance shots, so when you tally that up over a whole season, we’ll have “created” a higher xG.

Under a manager who doesn’t want you to take a low chance shot and plays the stats more the xG will end up being closer to what’s achieved.

So can we actually take that much from it when to totalling it up like that?
 
DJ said:
With this though, we take a lot of low chance shots, so when you tally that up over a whole season, we’ll have “created” a higher xG.

Under a manager who doesn’t want you to take a low chance shot and plays the stats more the xG will end up being closer to what’s achieved.

So can we actually take that much from it when to totalling it up like that?
Click to expand...

But a few of the low chance shots will go in because of a deflection / poor keeping etc. On average 1 in 10 shots with a xG of 0.1 will go in compared with 1 in 2 if the xG was 0.5 so I think it is fair to tally them up.

The xG itself has been calibrated to come up with the value allocated to each shot. Understat always predicts higher number of goals than Opta, so I'm not surprised to see more teams underachieving on xG
 
dragonm8 said:
Does a certain players ability to shoot affect XG?

Does it depend on who's taking the shot?
Click to expand...

No, which is a valid criticism and why these are guides / for interest only (although betting sites are using them too). It is essentially based on the average attacker against the average keeper at the level the database collects data (mainly Europe's top leagues I think).

This means you can get 'stat-busters' (a term used in fantasy football, for players who consistently over many seasons score more than their xG) - Jamie Vardy was one.
 
Matt Stevenson said:
I had a look as I didn't know. Using Understat data, we are actually second worse when comparing xG against goals scored. We have scored 52 but were expected to score 60.68. Everton seem to be doing very badly.

8 teams are overperforming (in blue), 12 underperforming (in white)

View attachment 14170

I'll look at defensive record and post this too
Click to expand...
That's crazy, I was going on memory but I don't recall us systematically underperforming. I take it all back!

Thanks for doing the legwork.
 
Thanks for the reply Matt.

So for example Lutons XG against Arsenals is based on both teams keepers, and both teams strikers and attackers being equal?

If that's the case, I can't see how XG works, without bias.
 
dragonm8 said:
Thanks for the reply Matt.

So for example Lutons XG against Arsenals is based on both teams keepers, and both teams strikers and attackers being equal?

If that's the case, I can't see how XG works, without bias.
Click to expand...

There will be bias, although the amount is hard to know, The simplest example is on penalties where all players will have about 79% chance to score, which is based on historical data (so maybe 7889 penalties out of 10,000 were scored).

But in reality a specialist penalty taker against an average keeper, must have a better chance than someone who misses a few vs Martinez. It might only be a couple of percentage points, but it will definitely be there.

It wouldn't surprise me if the top teams had an advantage of 0.1xG per game because of their better skill, and the original graph showing Arsenal, City, Spurs and Newcastle all outperforming xG would support this.

I think xG is informative, as long as you don't put too much trust in the numbers and understand the limitations - these would include penalties not given, crosses that are are slightly too far in front of the attacker in a great position, neither of which would show on xG.
 
Matt Stevenson said:
No, which is a valid criticism and why these are guides / for interest only (although betting sites are using them too). It is essentially based on the average attacker against the average keeper at the level the database collects data (mainly Europe's top leagues I think).

This means you can get 'stat-busters' (a term used in fantasy football, for players who consistently over many seasons score more than their xG) - Jamie Vardy was one.
Click to expand...
And in reverse, so is Christie
 
GaryRB said:
And in reverse, so is Christie
Click to expand...
:).

Although to be fair to him, his xG this season is only 1.93. This is from 40 attempts so he is taking a lot of low chance shots. He almost scored from one of these vs Arsenal (hitting the bar, although the Solanke "foul" would have stopped it)

Last season he had an xG of 2.75 and scored 1 goal. He had 33 goal attempts so had better chances, but his goal vs Leicester would have had a high xG (and he still scuffed it past the keeper!)
 
I have very much enjoyed watching parts of the Scottish Championship this season. I have followed Raith Rovers for almost as long as I have AFCB. The matches are very competitive, no VAR, not much playacting and the standard is good enough to hold interest. Watched the playoff between Airdrie and Partick last night. It was a great game. No stench of unfairness after almost every match.
 
You must log in or register to reply here.

;