Changing football manager

[T 190]

http://www.bbc.co.uk/news/magazine-23724517

[PurpleCanary 5,553]

[quote user="T"]http://www.bbc.co.uk/news/magazine-23724517[/quote]

 

Two key statements there need to be kept in mind:1) "Teams suffering an uncharacteristic slump in form will bounce back and return to

their normal long-term position in the league, regardless of whether they

replace their manager or not."Presumably in this case "uncharacteristic" has to be defined as in relation to that team''s performance over several years. But it makes no allowance for how many managers and of what quality that club may have had during that time. If a team has had a long-serving and highly competent manager who has then replaced by an idiot the subsequent decline under the new man cannot possibly be called "uncharacteristic".2) "A short term decline in performance is

not a good reason to be firing your manager".But who is to say, as above, what is short-term? I take it the same definition  for "uncharacteristic" applies. But some declines are not temporary. Some declines, left unattended without a change of manager, will continue. And this especially applies to clubs who need to perform or even out-perform their financial position to stay in a particular division.

[snake-eyes 13]

That''s a decent response Purple.

 

For me the perfect example of this is David Moyes at Everton.  He started well, before a blip in his second season. That was put to rights the next followed by a minor blip the following season (11th). After that the never finished outside the top 8.

 

The poor form was in the second half of the second season, but were safe from relegation several games before the end. By the End of September the following season they were 3rd with 16 pts.

 

This is what demonstrates a dip in form and not to remove the manager.

 

The question is can you clasify a run of 26 games 5 wins as a dip in form? or is this the norm?

 

As ''the Run'' created 6 wins and 4 draws you would have to say that as brilliant as it was, this is not the norm.  If it is a dip it is a very long one!

 

So for me whilst the report highlighted by T is very interesting and probably correct in its conclusions, I don''t believe Norwich fall into this catagory of uncharacteristic drop in form. Therefore this brings into play the viable posibility of changing manager.

 

I am not saying this is what should happen, but by all accounts it cannot be dismissed.

 

Snake

[PurpleCanary 5,553]

[quote user="snake-eyes"]

That''s a decent response Purple.

 

For me the perfect example of this is David Moyes at Everton.  He started well, before a blip in his second season. That was put to rights the next followed by a minor blip the following season (11th). After that the never finished outside the top 8.

 

The poor form was in the second half of the second season, but were safe from relegation several games before the end. By the End of September the following season they were 3rd with 16 pts.

 

This is what demonstrates a dip in form and not to remove the manager.

 

The question is can you clasify a run of 26 games 5 wins as a dip in form? or is this the norm?

 

As ''the Run'' created 6 wins and 4 draws you would have to say that as brilliant as it was, this is not the norm.  If it is a dip it is a very long one!

 

So for me whilst the report highlighted by T is very interesting and probably correct in its conclusions, I don''t believe Norwich fall into this catagory of uncharacteristic drop in form. Therefore this brings into play the viable posibility of changing manager.

 

I am not saying this is what should happen, but by all accounts it cannot be dismissed.

 

Snake

[/quote]

 

The Everton example is a good one. To give another, take Man Utd under Ferguson. Ie, a manager in place for many years at a club whose relative financial position doesn''t change much during that time. Then if he  has a season or half a season around tenth or eleventh rather than in the top three you can certainly call that "uncharacteristic". To sack Fergusn for a blip like that would be absurd.But some blips are not blips. They are the start of a slide. As happened, for example, at Man Utd after  Busby left. It took several years and a few managers before they got it right with Ferguson. Ditto Charlton post-Curbishley.The other point is that this research is based on the Dutch league. Now I am not an expert but it is a league tradionally dominated by a small numbers of clubs, such as Ajax, PSV and Feyenoord. Looking at the second tier there is only one club there (Sparta Rotterdam) I might have expected to see in the top flight. In other words I suspect there is less upward and downward mobility, and so less likelihood of the manager being able as crucial s factor as in other leagues, such as England, where you can currently find some "big" clubs (Forest, Sheffield Wednesday, Leeds, Birmingham) below the top tier.

[T 190]

I''d agree it does not necessarily mean you should not change the manager. The report does say that the findings also apply to the premier league. The question is whether an underprformance is tempoary or not and what constitutes an underperformance. Reports from other studies which my own analysis confirms is that teams largely perform  in line with their wage bill so in NCFC''s case the expected reversion to mean performance is currently 17th plus or minus 2 places. Consequently, CH outperformed last season and is performing in line with the expected performance this season which suggests changing manager will not have any benefit unless you can find a manger who outperforms as Lambert and CH have done.  This is entirely logical as the best players tend to be paid the best wages and if there is an overperformance then the manager/player end up moving so over a period of time a team naturally progresses to its mean performance.5 games is simply too short a period to revert to the mean performance.

 

Moyes is a good call as he is the one and only manager who has consistently outperformed the performance indicated by finances and even then not always. I think fans find this difficutlt to understand as statistics, probability theory and decision theory are not widely taught so there is a mistrust of something that people do not understand and also humans naturally do not grasp the reality as they use perception filters - an exaggerated focus on short-term performance being a classic example and also seeing one-offs such as Swansea and QPR as disproving the norm when actually these fit entirely with the normal probability distributiuon model i.e a few teams out and under perform from time to time but the probability is low over the long term and teams revert to their position indicated by finances. I strongly suspect McNally is informed of the Deloitte reports and other studies is and intelligent enough to appreciate this reality as CH and Lambert also do.  He also has a high financial incentive to ensure any change is reality in the clubs interests and will make a nard nosed objective decision rather than an emotionl subjective nalysis as many fans do. Managing fans expectations can have a real impact on the reality though and perhaps the club could do more to highlight the reality.

[PurpleCanary 5,553]

[quote user="T"]

The report does say that the findings also apply to the premier league. 

[/quote]

 

Since you place almost total store by hard statistics and have little time for guesswork I think we need to be clear about this. As I understand the article the only hard statistics available on this subject are those compiled from the Eredivisie. There is no mention of any other statistical evidence, from the Premier League or elsewhere. There is this:"While Ter Weel''s research focused on Dutch football, he argues that this finding is not specific to the Netherlands. Major football leagues in Europe, including England, Germany, Italy and Spain also bore out the same conclusion."Which is far to woolly to be relied upon or taken seriously. He argues that his finding is borne out elsewhere. Based on what? Anecdotal evidence? That won''t do. A general feeling that this is the case? Again, that won''t do. And there is the unscientific imprecison of "bore out the same conclusion". Borne out totally, or mainly, or just a bit? Borne out all the time or just some of the time?If this Dutch economist or someone else does a similar survey of the Premier League over several years then it may be possible to say what applies in the Eredivisie also applies here. But not until then.

[T 190]

PC I would agree that the wording is somewhat ambiguous but the comment that while focused on Netherland other leagues bore out the same conclusions does suggest that some review of other leagues was performed and that the findings also applies to the premier league. Other studies for the premier league have been reported as showing that performance is primarily driven by finance so it is hardly surprising and logical that changing manager does not have an impact. My own analysis for the past couple of seasons to check the findings is consistent with the thinking. To be honest I was surprised just how strong the correlation was between finance and performance. What is clear is that there is no evidence to suggest changing the manager of a team that is performing in line with expectations is beneficial. I think you will find a few Wolves fans that agree. In fact spending your money on changing manager will reduce the available player budget and therefore is more likely to have an adverse impact on performance.

 

PC you are surely a lawyer who has worked in the world of beyond resonable doubt as opposed to the wrold of reasonable probability. Its football not proof of the Higgs Boson.

[BOO 0]

Balance of probabilities !

[Reggie Strayshun 0]

Good discussion this. I guess it''s easy to forget that , unlike many industries, football is essentially financially controlled not by hardened professionals in that line of business, but by well meaning amateurs. Certainly in the case of the lower leagues, but in a surprising number of Prem outfits. That''s why the clubs like Norwich have got Mc Nally etc in to make up for the likes of the Wynn Jones'', albeit well meaning, lack of knowledge. Basically Abramovich knows sod-all about football, but he''ll decide on a manager''s fate, despite that

So, he who pays the piper calls the tune, ultimately. You can analyse stats till kingdom come, but, if the guy with the cash wants a new boss "to freshen things up a bit", then, in the end, he will get his way.

[PurpleCanary 5,553]

[quote user="T"]

PC I would agree that the wording is somewhat ambiguous but the comment that while focused on Netherland other leagues bore out the same conclusions does suggest that some review of other leagues was performed and that the findings also applies to the premier league. Other studies for the premier league have been reported as showing that performance is primarily driven by finance so it is hardly surprising and logical that changing manager does not have an impact. My own analysis for the past couple of seasons to check the findings is consistent with the thinking. To be honest I was surprised just how strong the correlation was between finance and performance. What is clear is that there is no evidence to suggest changing the manager of a team that is performing in line with expectations is beneficial. I think you will find a few Wolves fans that agree. In fact spending your money on changing manager will reduce the available player budget and therefore is more likely to have an adverse impact on performance.

 

PC you are surely a lawyer who has worked in the world of beyond resonable doubt as opposed to the wrold of reasonable probability. Its football not proof of the Higgs Boson.

[/quote]

 

T, you might want to believe that is what it means but it doesn''t say that. And crucially there is no reference to any other study. You cannot have it both ways. If you want to rely on hard statistics from a serious survey, and rubbish posters who rely on feelings and supposition and anecdotal evidence, then fine. But you can''t then now start throwing into the mix your own supposition, individual cases (such as Wolves) and wishful thinking based on surveys you would like to have been done but seem not to exist. If such a survey as this Dutch one had been carried out for the Premier League you would have quoted it by now. It would have been all over the media. It would be brought up every time there was a managerial upheaval.Until such a survey is done, or discovered, it is not possible to say with certainty that changing the manager in the Premier League is as irrelevant as it has been shown to be (with some serious caveats) in The Netherlands.

[A Gay Schoolboy 0]

Its interesting that you are saying that 5 games isn''t enough time to judge a manager, as I''m reading the actual paper mentioned in the article and it uses only the 4 games prior to the sacking, and then the new manager''s first 4 games as its analysis periods.

[T 190]

Thank you Wolf!

PC - You are being somewhat pedantic - I''m not saying it is definitive but it is more substantial than pure speculation. They is also indirect proof as I''ve mentioned otherwise the link between finance and perfoance would not be so strong. It certainly supports the argument for not changing manager as opposed to changing manager. There is certainly no evidence I''ve seen to suggest changing manager helps. I think you will find that the LMA objects every time a maanger is changed on the basis that it takes time for a manager to change things around and it is expensive. I will give you credit though for picking up the ironic reference to Wolves though! Now if you want to do somehting useful go away and read the latest Deloittes report and the Numbers Game.

[A Gay Schoolboy 0]

I don''t know why you are talking about the link between finances and position in relation to this particular issue. The paper''s authors explicitly rule out using budget as a metric for their baseline performance measure, as in the short term (i.e. a season or less) they claim the link, and I quote, is "unclear" and "far less than perfect".

[Reggie Strayshun 0]

Dandy,I can see why the paper might want to exclude the link performance v finances, but, as I was saying in my earlier post, in reality, in many cases, it''s just not possible to do that.

If the club owner has pumped several million quid into it, and he suspects that the manager is not spending it wisely, I think it''s fair to say that he''ll have a view on it. And, as I said, ultimately, he will be the one with the final decision as to whether the manager staays or walks, surely ?

[A Gay Schoolboy 0]

[quote user="Reggie Strayshun"]Dandy,I can see why the paper might want to exclude the link performance v finances, but, as I was saying in my earlier post, in reality, in many cases, it''s just not possible to do that.

If the club owner has pumped several million quid into it, and he suspects that the manager is not spending it wisely, I think it''s fair to say that he''ll have a view on it. And, as I said, ultimately, he will be the one with the final decision as to whether the manager staays or walks, surely ?[/quote]You misunderstand Reggie. The motivations of the sacking aren''t what is being studied. In order to measure the relative performance of a club before and after they hit a short term dip in form considered "sackable", you need a baseline metric of how that squad *should* have been performing, and how they should be expected to perform in future. T believes this metric should be based on financials, while Dr Bas ter Weel et al, the men who''s work T is appropriating (apparently without having actually read it) disagrees.