Match Thread [FT] Rangers 0 - 2 Hibernian

[Rousseau 11,608]

1 hour ago, Tannochsidebear said:

One thing I am not is arrogant, apologies if it comes over that way. Pretty sure im in the minority and sounding like a dinosaur for not buying into these new-fangled stats side of football discussions, but when there are such clear errors in the data they make a decent discussion point do they not? I dont mind xG as a discussion point, but it is being accepted as facts, when it is quite clearly not facts. 

Ok, apologies. 

 

Where's the error? It's based on what has actually happened.

 

I don't know what you mean by 'fact.' It's just the probability. And it is just a discussion point. 

 

Let's simplify it and just consider penalties. From the historical data, penalties are scored 79% of the time. You can watch them all back. It gets a value of 0.79 xG.

 

The only fact is that a penalty has been scored 79% of the time in the past. That's indisputable, surely?

 

We then analyse the data by comparing it to what actually happens. (It's the variance that is interesting, as it can be analysed: Are the players rubbish or just unlucky? Are the chances poor quality? Are we not creating enough? Is a player scoring less or more than he should be? Who is getting chances from high xG situations?)

 

A one-off penalty won't tell you much. If after 10 penalties and the player has 2 goals from an xG of 7.9 (10 * 0.79), then he's rubbish.

 

As I've said, the Opta model goes into much, much more detail and volume than what anyone could ever judge with their own eyes. 

 

Have a browse of this article, which will explain it better than I ever could: https://theanalyst.com/2023/08/what-is-expected-goals-xg 

[Rousseau 11,608]

3 hours ago, Tannochsidebear said:

What i do say is that Dessers misses a bucketful of chances regularly, and if he was more clinical, we might not be miles behind domestically. There was a suggestion that because he is matching this mythical xG that he is performing well for the team, which I disagree with.  

I agree with you. I think Dessers is wasteful. 

 

The xG does not show he is performing 'well'. It shows he is performing slightly below average (a 0.52 difference is not insignificant). It's just that he's not as wasteful as you or I thought he was. 

 

Combining the data with what I see, I would suggest that he tends to miss easy chances and score harder chances (e.g. Dundee Utd), which would skew the numbers slightly. 

 

He does only average 2-3 shots a game (64 shots he has taken this season) so his misses are more obvious. 

 

At the end of the day, 12 goals in 32 league games from a number 9 is not great. It's also more of a problem because no one else is scoring - Dessers is hitting the same numbers, or better, than Morelos (excluding the fact that Morelos was a superior player in all other regards, IMO). 

 

It's interesting that you see the variance between the data and what you see and think, 'the data is bollocks'; whereas I see the variance and think, 'let's dig deeper into the context to try explain it.' 

[Blue Moon 1,587]

5 hours ago, JohnMc said:

Thanks, that's really helpful. So our XG goes up if we create 'better' chances, not if we change centre forward? But a more clinical centre forward should score more goals? I don't have a problem accepting both those statements as true. 

 

So when someone says that Dessers misses a lot of chances it's true, it's just that they're not considered 'easy' chances by whoever compiles this data. 

My main criticism of Dessers earlier in the season was his work outside the box. I felt his hold-up play and link-up play was poor and so if he wasn't scoring he wasn't contributing anything. That side of his game has improved, his all round contribution is better now. 

Morelos would beat him hands down in taking the ball with his back to goal and redistributing it.  Dessers doesn't do enough of that for me.

[JohnMc 3,157]

Yeah, peak Morelos was so good at that. I think that's why others scored so much more then too, he occupied at least one, often 2 defenders, was still able to either hold it up, lay it off or draw a foul allowing Kent, Aribo, Arfield, Roofe etc to use the space he'd created. Igamane has shown moments that he might be able to do this, but he's not there yet. 

[Blue Moon 1,587]

5 minutes ago, JohnMc said:

Yeah, peak Morelos was so good at that. I think that's why others scored so much more then too, he occupied at least one, often 2 defenders, was still able to either hold it up, lay it off or draw a foul allowing Kent, Aribo, Arfield, Roofe etc to use the space he'd created. Igamane has shown moments that he might be able to do this, but he's not there yet. 

I agree about Igamane.  He looks like he has the potential to be a great player and with a better temperament than Morelos.