Jump to content
Sign in to follow this  
Herman

Guardian Preview.

Recommended Posts

12 minutes ago, Canary Wundaboy said:

For once, something on that rag worth reading. Thanks for the share, good piece. 

What would you prefer? The Sport?

 

There’s a great article in the Guardian today about baked potatoes, so that’s two things.

  • Like 1
  • Haha 2

Share this post


Link to post
Share on other sites

The article says that the stats say we scored more goals than we should have done.  That kind of makes you wonder if stats are of any help......with talk of a "return to the mean", as if that means anything other than that we have something special going for us which stats can't explain.

The achievement is a statistical anomily and that makes me smile - because we have something else - it's indefineable, it's a touch of gold - it's that we are more than the sum of our parts and there is no reason why that cannot continue - and the statisticians have no answer to it.  

I would say the experience gained from last season will have galvanized every player at the club to improve and step up - and again that is the unknown factor that stats cannot predict - that of how the success will affect the players individually and collectively.  Farke is a clever operator and knows how to get the mentality right - he is a Brian Clough like influence (without the eccentricity) - everyone loves him from top to bottom of the club and he has the quality players he needs and the ability to get the best out of them. 

The prognosis is good - and as much as all these articles try and get a handle on how we will do, they cannot quite believe we will do that well......but we do have the X factor, Daniel Farke, who with Webber have already transformed the football, creating real success from practically nothing and who will continue on the same path.....

......upwards.

 

Share this post


Link to post
Share on other sites
25 minutes ago, Herman said:

A rather decent season preview in today's paper if you fancy a quick read.

https://www.theguardian.com/football/2019/aug/06/premier-league-201920-preview-no-14-norwich-city

Not a bad preview, but this over reliance on statistical **** is a bit tiresome. We scored 20 more goals than expected - well, erm, expected by whom?

Considering most of the squad players were either new to the division and/or extremely young and inexperienced, it is difficult to understand why an effectively invented number would be taken as Gospel truth that we 'overachieved' last season.

Other than that, at least it shows a little more research than most of the other previews in which we've featured. 

Share this post


Link to post
Share on other sites

A very good preview by somebody who appears to have actually looked under the bonnet.

Share this post


Link to post
Share on other sites

It means that based on the quality of the chances we created we scored twenty more goals than would be expected. So our finishing was pretty clinical. We need that to continue as it's likely we will get less chances in the Prem.

  • Thanks 1

Share this post


Link to post
Share on other sites
13 minutes ago, Ian said:

Not a bad preview, but this over reliance on statistical **** is a bit tiresome. We scored 20 more goals than expected - well, erm, expected by whom?

Considering most of the squad players were either new to the division and/or extremely young and inexperienced, it is difficult to understand why an effectively invented number would be taken as Gospel truth that we 'overachieved' last season.

Other than that, at least it shows a little more research than most of the other previews in which we've featured. 

Agreed, it’s absolute nonsense. I’m sure someone will come along and claim otherwise but football just doesn’t work like that

Share this post


Link to post
Share on other sites

I occasionally watched motd last season and they had an ‘expected goals’ stat, or something similar....which does seem to be a bit of a pointless thing.  There does seem to be a bit of a ‘reliance’ on stats nowadays but some seem to mean little - you can play well and have 10 decent chances against a keeper playing out of his skin and score none, or get one half chance that skews in off a defender....what do the ‘stats’ say about those ones?

Share this post


Link to post
Share on other sites
27 minutes ago, Hairy Canary said:

It means that based on the quality of the chances we created we scored twenty more goals than would be expected. So our finishing was pretty clinical. We need that to continue as it's likely we will get less chances in the Prem.

Yes, and saying our finishing was, in general, clinical is a perfectly reasonable statement.

However, suggesting that means we overachieved is a nonsense - we simply have quality goalscorers in the side.

  • Like 1

Share this post


Link to post
Share on other sites

I like the concept of the xpg but it doesnt seem to actually match the results the majority of the time.

Share this post


Link to post
Share on other sites

There used to be a poster on here called westcoast canary who enjoyed quoting the stats, although he seemed to assume that because our goals scored was above xG this could be expected to "revert to the mean", whereas because our goals conceded was also above the xG stat, this meant we were "defensively weak".  Hmmm.

 

Interestingly though, the excess of our goals scored above xG declined in the 2nd half of the season, which is positive in that we were basically overperforming, for some reason, in the first half, and not so much in the 2nd half of the season.

 

The article is interesting although I'm not sure it came up with much that hasn't already been covered on here at length.

 

I have TBH Fahrmann has looked a bit twitchy in goal for me.  He may settle and replace Krul in time, but I'd definitely rather have Krul playing on Friday.

Share this post


Link to post
Share on other sites

I wonder what the 'xpg' would of been had we taken our penalties? This may be be one statistic too far.

Stats are good for guidance but they are not God.

  • Like 1

Share this post


Link to post
Share on other sites
49 minutes ago, Ian said:

Yes, and saying our finishing was, in general, clinical is a perfectly reasonable statement.

However, suggesting that means we overachieved is a nonsense - we simply have quality goalscorers in the side.

Yep totally agree. Clinical finishing is probably the most important part of the game and we won the league because we were pretty much the best at doing it.

Share this post


Link to post
Share on other sites
22 minutes ago, Yellow Fever said:

I wonder what the 'xpg' would of been had we taken our penalties? This may be be one statistic too far.

Stats are good for guidance but they are not God.

Very true but statistically we (massively) underperformed from penalties and overperformed from open play, the 20 goal excess is the net effect.

Share this post


Link to post
Share on other sites

Does xpg work the same way as when you see a bloke who’s a bit tubby and thinning on top and clearly his better days are a distant line on the horizon. then you see his misses and you’re thinking how the **** has he managed that!? He’s far outperformed his expected pulling ratio (epr)??

  • Haha 1

Share this post


Link to post
Share on other sites
2 minutes ago, Jobsworth Canary said:

Left wing rag not worth listening to . 

Do you always listen to paper?

  • Haha 3

Share this post


Link to post
Share on other sites

It’s a good preview and the stats are there for context only. 

Expected goals suffers as it is a bit of a silly name dreamt up by statistician. However, what it shows and how it is calculated is perfectly reasonable. All Norwich fans know that the finishing last season was exceptional and the team scored a lot of long range shots - more goals from outside the box than any other team. 

It is a perfectly valid question to ask, “can Norwich maintain this?”  

All successful teams will out score their expected goals across a season. Leicester did so massively when they won the league. It is rare however for a team to do so season after season after season. 

My biggest doubt for whether Norwich can survive next season is borne from the high level of ‘over achievement’ in finishing. 

  • Like 1

Share this post


Link to post
Share on other sites
1 hour ago, hogesar said:

I like the concept of the xpg but it doesnt seem to actually match the results the majority of the time.

In a single game, it generally doesn’t but when mapped across a larger sample size it does tend to match up pretty well. 

Share this post


Link to post
Share on other sites
10 minutes ago, Bethnal Yellow and Green said:

In a single game, it generally doesn’t but when mapped across a larger sample size it does tend to match up pretty well. 

Yeah, understood. But you can see why its rubbished quite quickly especially by those who dont like the idea of stats; motd showing the xpg as 1-1 when Bournemouth win 4-0 makes it look silly on the face of it.

Share this post


Link to post
Share on other sites
33 minutes ago, hogesar said:

Yeah, understood. But you can see why its rubbished quite quickly especially by those who dont like the idea of stats; motd showing the xpg as 1-1 when Bournemouth win 4-0 makes it look silly on the face of it.

If taken in isolation a stat is pretty much always useless. There will always be anomalies that are easy to use to disprove something. Doesn't mean that it does. 

  • Like 1

Share this post


Link to post
Share on other sites

49 minutes ago, Bethnal Yellow and Green said:

In a single game, it generally doesn’t but when mapped across a larger sample size it does tend to match up pretty well. 

My understanding is that expected goals doesn't take into consideration defensive positioning or quality of attacker. I may well be out of date with my information, but assuming this is accurate, by not distinguishing between Pukki and Zimbo (no offence Christoph) having a pop from 20 yards it is fairly silly to see it as anything other than a moderately interesting discussion point. 

I also don't really understand why it would be considered to have any predictive ability whatsoever, especially when considering the step up in terms of quality between Champs and Prem. Probably more accurate than suggesting we overachieved last season, would be to state that Norwich had a greater number of quality finishers upfront and in midfield, and therefore made more of the chances that were generated. 

Let's be honest; if a model is suggesting that over the course of 46 games we scored nearly 25% more goals than "expected" it seems obvious to say the model does not have much connection to reality.

Share this post


Link to post
Share on other sites
11 minutes ago, Ian said:

My understanding is that expected goals doesn't take into consideration defensive positioning or quality of attacker. I may well be out of date with my information, but assuming this is accurate, by not distinguishing between Pukki and Zimbo (no offence Christoph) having a pop from 20 yards it is fairly silly to see it as anything other than a moderately interesting discussion point. 

I also don't really understand why it would be considered to have any predictive ability whatsoever, especially when considering the step up in terms of quality between Champs and Prem. Probably more accurate than suggesting we overachieved last season, would be to state that Norwich had a greater number of quality finishers upfront and in midfield, and therefore made more of the chances that were generated. 

Let's be honest; if a model is suggesting that over the course of 46 games we scored nearly 25% more goals than "expected" it seems obvious to say the model does not have much connection to reality.

The better models take into account position of defenders among many other variables. 

The point of them is to not take into account who is taking the shot, as it is suppose to be an average. So good attackers will always be slightly over expected.

Where it is useful (and remember it was invented by professional gamblers to help them make money) is to identify where a team or player are maybe on a bit of a ‘hot streak’ and scoring well above (or below) what would be considered the normal variances. 

Expected goals is taking shot data to the next level. Instinctively we know that when one team takes 20 shots and other only has 1, that should the team with 1 shot win, it would feel a bit of a smash and grab. However, if the one shot the team took was from a penalty and the 20 the other team took were from the half way line - then expected goals would reflect that, giving a more accurate picture of the game. 

The acid test for these kind of stats is ‘who uses it and how is it used’ and expected goals is both used by football teams who spend a lot of money to buy this kind of data and by professional gamblers who use it to help them make money. If it didn’t mean anything, it wouldn’t be used. 

Share this post


Link to post
Share on other sites
27 minutes ago, Bethnal Yellow and Green said:

The better models take into account position of defenders among many other variables. 

The point of them is to not take into account who is taking the shot, as it is suppose to be an average. So good attackers will always be slightly over expected.

Where it is useful (and remember it was invented by professional gamblers to help them make money) is to identify where a team or player are maybe on a bit of a ‘hot streak’ and scoring well above (or below) what would be considered the normal variances. 

Expected goals is taking shot data to the next level. Instinctively we know that when one team takes 20 shots and other only has 1, that should the team with 1 shot win, it would feel a bit of a smash and grab. However, if the one shot the team took was from a penalty and the 20 the other team took were from the half way line - then expected goals would reflect that, giving a more accurate picture of the game. 

The acid test for these kind of stats is ‘who uses it and how is it used’ and expected goals is both used by football teams who spend a lot of money to buy this kind of data and by professional gamblers who use it to help them make money. If it didn’t mean anything, it wouldn’t be used. 

I would be surprised if many professional gamblers rely on an expected goals model as opposed to arbitrage and other such betting methods, but I will take your word for it.

You've hit the nail on the head with regards to the crux of my issue with it however, and that is application. If you are using a decent model that shows that a team has scored a lot of difficult chances in recent games, as PART of an overall betting strategy which takes into account other factors, such as injuries to key players, quality of opposition and so on, then I assume it could have some use in such a limited scenario. I'm also sure gamblers would use any statistic they think gives them an edge, and am acutely aware of how the real pros manipulate mug punters into placing silly bets by using the latest fad statistic.

However, I'm sure we can agree that using it to try and predict a team's upcoming results and/or to reach the conclusion that Norwich have over-achieved in scoring so many goals is just a bit daft when there are so many other variables in play.

To put it another way; you have said yourself that the best attackers will always be above average expected goals - so how can determine whether Norwich had a "bit of a hot streak" last season, or whether we simply have a far higher quality of attacker than anybody thought? I suspect next season will answer that question more accurately than a single basic statistical model...

  • Like 1

Share this post


Link to post
Share on other sites

I liked this piece, it painted a realistic view of what we are about and what our strong points and flaws are.

I must say I block out the dross about all the stats and the predicted goals for and actual goals etc...while I'm reading.

Stats can be valuable , but nowadays it seems it's often more about the stats than the actual team , coach, setup and players.

We might have overachieved last season, but that was down to good management , players stepping up when they got their chances , ... 

Share this post


Link to post
Share on other sites
2 hours ago, ROBFLECK said:

I liked this piece, it painted a realistic view of what we are about and what our strong points and flaws are.

I must say I block out the dross about all the stats and the predicted goals for and actual goals etc...while I'm reading.

Stats can be valuable , but nowadays it seems it's often more about the stats than the actual team , coach, setup and players.

We might have overachieved last season, but that was down to good management , players stepping up when they got their chances , ... 

Agree - The central point it did make which is clearly true is that Pukki scored 30 odd goals. Can he do something equivalent in the EPL or will others step up. If he/we don't then we could be in trouble! 

 

Good balanced article - and they predict we survive (just) 😀

Share this post


Link to post
Share on other sites
5 minutes ago, Yellow Fever said:

Agree - The central point it did make which is clearly true is that Pukki scored 30 odd goals. Can he do something equivalent in the EPL or will others step up. If he/we don't then we could be in trouble! 

 

Good balanced article - and they predict we survive (just) 😀

Depends what you mean by equivalent. We won't need to score 90 odd goals and finish top for this season to be successful.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...