Eddie Howe

[Optimistic Nut]

9 minutes ago, Cronky said:

 

Lies, damned lies and statistics. More than anything, for me, this calls into question the value of this concept of 'expected goals', or how it's calculated.

 

Man U on 44 points, us 8 points ahead of Spurs and 10 in front of Villa? 

 

The best stat that I've come across for predicting results is 'shots on target'. 

 

Using this argument then, what's a more likely chance to score. A 30-yard daisy cutter that rolls straight to the keeper, or a chance like Joelinton's yesterday?

 

 

Edited May 20 by Optimistic Nut

[Boey_Jarton]

15 minutes ago, Cronky said:

 

Lies, damned lies and statistics. More than anything, for me, this calls into question the value of this concept of 'expected goals', or how it's calculated.

 

Man U on 44 points, us 8 points ahead of Spurs and 10 in front of Villa? 

 

The best stat that I've come across for predicting results is 'shots on target'. 

Nearly everything about the table reconciled with my intuition. 

[Andy]

There isn't a stat that can predict results, at least not very well, football is unpredictable.

 

This is proven every time someone posts a table like that to show a team has underperformed or overperformed their xG/xPts or whatever else. It's not that the team has under or over performed, it's that the data is flawed apart from at the extreme ends of the scale. A team's results over a full season is their correct level. 

[madras]

21 minutes ago, Cronky said:

 

Lies, damned lies and statistics. More than anything, for me, this calls into question the value of this concept of 'expected goals', or how it's calculated.

 

Man U on 44 points, us 8 points ahead of Spurs and 10 in front of Villa? 

 

The best stat that I've come across for predicting results is 'shots on target'. 

Shots on target would say that a shot that trickles towards the goal from 40yds that the keeper literally puts his hat on is as good a chance as someone missing an open goal from 2 yds out.

 

XG is a guide, not a conclusive picture.

 

XG is just like coming out from the match and saying to your mate "we got lucky there, we won one nowt but they missed three sitters"

[The College Dropout]

16 minutes ago, Andy said:

There isn't a stat that can predict results, at least not very well, football is unpredictable.

 

This is proven every time someone posts a table like that to show a team has underperformed or overperformed their xG/xPts or whatever else. It's not that the team has under or over performed, it's that the data is flawed apart from at the extreme ends of the scale. A team's results over a full season is their correct level. 

Don’t agree with this. 
 

Over the course of a season the stats tell a true story. But you might have to dig to find the true story. 
 

For example Dyche Burnley consistently beat their Xgc. A combination of high volume low quality shots conceded and a great GK was the reason. 

[The College Dropout]

7 hours ago, Pata said:

Absolutely mad looking table in expected points considering the circumstances, finished fourth ahead of Chelsea. N-O darling Emery 10 points behind, media darling Ange 8 points behind. Not even going to mention Erik the fraud, hope he gets another season. Eddie is a magician.

 

Villa have been doing that since Emery joined. Think Martinez is a big part of that.  
 

I think selling Martinez has been Arteta’s biggest mistake. He might’ve made the difference this season or last. 

[Bondedcrown]

8 hours ago, AyeDubbleYoo said:

You want to be proven wrong but you think he’s one of the best managers in the world? 

Some of the quotes on here are embarassing! a lack of having played anything other than 5 a side with their mates being the main cause!

[Bondedcrown]

2 hours ago, Monkey Alan said:

I like how he's not one of these managers who tells his players do it this way or that way. He goes 'do it this way, if you like'.

 

 

 

You know that how?

[Andy]

37 minutes ago, The College Dropout said:

Don’t agree with this. 
 

Over the course of a season the stats tell a true story. But you might have to dig to find the true story. 
 

For example Dyche Burnley consistently beat their Xgc. A combination of high volume low quality shots conceded and a great GK was the reason. 

 

They might "tell a story" but they aren't a good metric for evaluation of where a team should have finished, or whether a team should have won a game. 

 

In the case of Dyche's Burnley for example, they didn't "outperform their xGc", they performed at their expected level because they have a good keeper and restricted chances. 

 

These stats don't take into account individual ability of players at both ends of the pitch, which is the major flaw. A chance is only as good as the players involved. 

[80]

I'm under the impression the publicly published xGs are a pale shadow of what clubs calculate internally.

[Cronky]

1 hour ago, Optimistic Nut said:

 

Using this argument then, what's a more likely chance to score. A 30-yard daisy cutter that rolls straight to the keeper, or a chance like Joelinton's yesterday?

 

 

 

 

With respect, that's false logic. The stat works because a shot on target is a statistical indication of a good scoring chance. A shot off target is an indication of a poorer chance. The whole picture is that it's not so much the number of chances as a whole that a team creates, but the number of good chances. 

 

You might get a naive manager who looks at the 'shot on target' stat and decides to instruct his players to hit the ball towards the opposition goal from all areas of the field. It wouldn't work because those situations do not represent 'good chances'. 

 

What makes football results hard to predict is that a team with a lot of possession who is trying to break down a packed defence might not get its due reward because of the value of a counter attack. A team is more likely to score when it's 2 on 2 rather than 9 on 9. 

 

There was a period - I think it was the 80s - when someone (Charles Hughes?) produced a stat that supposedly showed that the fewer the number of passes at team made in an attack, the more likely they were to score. This persuaded some teams to just hoof the ball forward rather aimlessly, no matter what the situation. I think the 'fewer passes' stat was just an indication of the relative potency of the counter-attack.

[Rod]

5 minutes ago, Cronky said:

 

With respect, that's false logic. The stat works because a shot on target is a statistical indication of a good scoring chance. A shot off target is an indication of a poorer chance. The whole picture is that it's not so much the number of chances as a whole that a team creates, but the number of good chances. 

 

You might get a naive manager who looks at the 'shot on target' stat and decides to instruct his players to hit the ball towards the opposition goal from all areas of the field. It wouldn't work because those situations do not represent 'good chances'. 

 

What makes football results hard to predict is that a team with a lot of possession who is trying to break down a packed defence might not get its due reward because of the value of a counter attack. A team is more likely to score when it's 2 on 2 rather than 9 on 9. 

 

There was a period - I think it was the 80s - when someone (Charles Hughes?) produced a stat that supposedly showed that the fewer the number of passes at team made in an attack, the more likely they were to score. This persuaded some teams to just hoof the ball forward rather aimlessly, no matter what the situation. I think the 'fewer passes' stat was just an indication of the relative potency of the counter-attack.

It was called POMO.  Position of maximum opportunity.  An awful theory that Sam Allardyce endorsed.  Just lump the ball into the box.

[Yorkie]

Just realised that the only sending off we had this season, in 51 games, was that daft one Gordon got late in the West Ham game. That's some pretty excellent levels of discipline; obviously drilled into the players to avoid suspension at all costs given the injury situation. 

 

For comparison, Everton and Man Utd also got one each, and Luton got none.

[Checko]

23 minutes ago, Cronky said:

 

With respect, that's false logic. The stat works because a shot on target is a statistical indication of a good scoring chance. A shot off target is an indication of a poorer chance. The whole picture is that it's not so much the number of chances as a whole that a team creates, but the number of good chances. 

(...)

But isn't this basically the point of xG? It's to analyse each individual shot and say, regardless of what the shooter actually did, was it a statistically good goal scoring chance? Rather than just relying on a cruder law of averages.

 

xG doesn't say who deserved to win on overall play (a particularly good GK should mean you outperform your xG against, a particularly clinical striker means you would outperform your xG). A defensive howler could gift you a gilt-edged chance or a penalty out of nothing that gives you extra xG the quality of your play didn't deserve.

 

xG, as far as I understand it, just tells you: based on the shots this team took, this is how many goals we would expect them to score. There's useful information to get from that, but also limitations. Just like any individual statistic.

[The College Dropout]

1 hour ago, Andy said:

 

They might "tell a story" but they aren't a good metric for evaluation of where a team should have finished, or whether a team should have won a game. 

 

In the case of Dyche's Burnley for example, they didn't "outperform their xGc", they performed at their expected level because they have a good keeper and restricted chances. 

 

These stats don't take into account individual ability of players at both ends of the pitch, which is the major flaw. A chance is only as good as the players involved. 

You use the underlying data with the outcome. The best pure finishers (Son, Kane) consistently outperform their Xg but their Xg is still useful. The better their Xg the better their outcome.

 

Burnley had some close scares relegation-wise and eventually went down. Because that approach is high-risk. You are right - the conclusion isn't that Burnley should have been relegated those years. They deserved to stay up for the reasons listed. But it did indicate they would be susceptible in the future if those narrow margins shifted.

 

FWIW I use stats extensively in FPL and I've done way better than most. I suspect that 98% of the people that finished above me use stats a lot too.

[Ginola14]

some pretty wild takes from MU fans here

 

https://www.redcafe.net/threads/how-good-is-eddie-howe.474692/page-10

 

to be fair I'd say most football fans are pretty clueless when it comes to other clubs and spout a lot of shite 

[PauloGeordio]

9 minutes ago, Ginola14 said:

some pretty wild takes from MU fans here

 

https://www.redcafe.net/threads/how-good-is-eddie-howe.474692/page-10

 

to be fair I'd say most football fans are pretty clueless when it comes to other clubs and spout a lot of shite 

Some sensible stuff in there praising us and Howe too. 

 

 

Edited May 20 by PauloGeordio

[Groundhog63]

7 minutes ago, PauloGeordio said:

Some sensible stuff in there praising us and Howe too. 

 

 

 

Yeah, further you dived in the more realistic the posts

[jackyboy]

This stays business is getting worse and really offers very limited information. When is a shot not a shot. High possession percentage does  not indicate control of a game

 

 

[timeEd32]

5 hours ago, The College Dropout said:

You use the underlying data with the outcome. The best pure finishers (Son, Kane) consistently outperform their Xg but their Xg is still useful. The better their Xg the better their outcome.

 

Burnley had some close scares relegation-wise and eventually went down. Because that approach is high-risk. You are right - the conclusion isn't that Burnley should have been relegated those years. They deserved to stay up for the reasons listed. But it did indicate they would be susceptible in the future if those narrow margins shifted.

 

FWIW I use stats extensively in FPL and I've done way better than most. I suspect that 98% of the people that finished above me use stats a lot too.

 

This is the key and you can use Man United to see how it can be valuable.

 

In 2022/23 Man United finished 3rd and won the League Cup. They were also third in the league in SOT. All hail ETH. But there were concerning signs like a GD that was more in line with 6th-8th and an xGA that said they were fortunate. Their xPTS were 6th, almost 9 less than what they actually earned, and they were trending down over the last 2.5 months of the season.

 

You could have argued heading into this season they underperformed their xG by a pretty large amount and so any defensive issues would be balanced out by more goals (which, to some extent, is what happened to us). But, instead, they created less chances and conceded more goals as the underlying stats said they should. Despite a worse defensive record they once again outperformed their xGA by a ludicrous amount this season, which is why their xPTS have them closer to relegation than top 10. Injuries definitely account for some of this, but two seasons of this trend should be highly concerning. And it's not like you watch them and come away impressed by a world-class GK and sublime counter-attacking football. These stats match the eye test far more than a 3rd place finish and then what was looking like 6th until a month or two ago.

 

[soyt]

Thank you Eddie for this season.

It had some great moments, some gut wrenching moments, a lot of bad luck and some incredible results.
I think he will be better manager after all this and i have no doubt the club have learned a lot about how to cope with European campaigns and injuries etc. Like some others, i was a bit frustrated at Eddie at some points. Feels kind of stupid now, but it is heat of the moment and all the rest.. I had to take some time off from this topic as it went round and round. But in the end this is clear: Even with the historic injury record & the horror cup draws, he took us to 7th and possibly to Europe, had good cup runs and conducted himself brilliantly at every point. 

I can't wait for next season, it is going to be great!

[ATB]

There you go, my favorit person again 

[Jack27]

In Howe we trust. Another legendary win over Spurs

[PauloGeordio]

Despite suggestions that Newcastle are on the verge of signing Lloyd Kelly and Tosin Adarabioyo, players the club like, Howe stated that he was not aware of any imminent announcement.

Howe said: "No, nothing close. I am a step removed from everything because we are out here, time difference plays a part. I am not fully up to date. But I don't think anything is close."
 

[sempuki]

1 minute ago, PauloGeordio said:

Despite suggestions that Newcastle are on the verge of signing Lloyd Kelly and Tosin Adarabioyo, players the club like, Howe stated that he was not aware of any imminent announcement.

Howe said: "No, nothing close. I am a step removed from everything because we are out here, time difference plays a part. I am not fully up to date. But I don't think anything is close."
 

Can ignore that. He's a bit of a fibber. He'll know exactly what's going on.