Is there a “stats based approach” already in place at Cardiff City?

CoymayMaybe the afternoon on which City CEO Ken Choo has had to issue a statement on the club’s website outlining his disappointment with a journalist at Wales Online following a story on the weekend which headlined that Mr Choo believed Russell Slade had “failed” as Cardiff City manager is not the best time to place too much faith in another article that media outlet had published on the club in the past few days.

However, I’m going to accept as true something they reported on Friday when they stated that they understood that the new Head Coach would be “young, energetic, stats driven and very much in tune with modern-day football”.

Reading that piece again, the strong impression given is that Wales Online were being briefed by someone in a senior position at the club. It seems reasonable to me to assume that this person carries some influence and the lack of any subsequent denial from the club about the claims made in the piece could be viewed as an implicit approval of what was said.

Therefore, if we accept that the new man is indeed going to be ”young, energetic, stats driven and very much in tune with modern-day football”, what does that tell us? Well, we are talking about a profession where someone in their forties is often described as young and energetic can mean all sorts of things really – will he take part in practice matches in training? Will he be unable to keep still for more than five seconds in the dugout? etc. etc.

Similarly, “very much in tune with modern-day football” is sufficiently vague to mean all sorts of things, but “stats driven”? No, for me, stats driven can only mean that we are talking about someone who sets great store by statistical information – this qualification seems perfectly clear to me and is the sort of thing which would not have been mentioned unless it was exactly what the club was looking for.

Now, there are questions of degree here which need to be taken into account. For example, I’ve seen Russell Slade described as something of a stats man in the past and he did make fairly common reference to what the stats said about the game just finished in his post match press conferences. The thing is though that, if the club were only looking for a continuation of the Slade approach to stats, why, seemingly, brief the local press that this will be a requirement? No, to me, this means that City want to go well beyond the influence that stats played in the Russell Slade era.

Certainly, “very much in tune with modern-day football” could relate to the increasing part statistical analysis is playing in the game, be it analysis of a player’s physical output during a game or season or analysis of their effectiveness in terms of things like anticipation and positioning, or overall influence on a match.

Rasmus Ankersen, the man behind the statistical analysis approach favoured by Midtjylland and Brentford, does he have some disciples at Cardiff City Stadium?

Rasmus Ankersen, the man behind the statistical analysis approach favoured by Midtjylland and Brentford, does he have some disciples at Cardiff City Stadium?

Probably, the most extreme example of reliance on statistical data comes from Denmark. FC Midtjylland are owned by the man who bought Brentford, Matthew Benham, and this fascinating article   explains how data is now king at the team which won the Danish Championship last season and beat Manchester United in the First Leg of the Europa League tie between the clubs in February.

The signing of Tim Sparv as described in the article is pretty mind boggling to this sixty year old. Around a decade ago, I can remember Sparv being at Southampton I think it was (I only knew this because I can recall scouting him and dismissing him as not being good enough for my Cardiff team while playing Football Manager!). Since then he has played in Sweden, Finland, the Netherlands and Germany for a variety of teams you’ll probably never see competing in the Group stages of the Champions League before arriving at Midtjylland because the data had marked him out as one of most important players in a Greuther Fürth side that had been identified as big over achievers in European club football in the model the Danes relied on.

Hardly surprisingly, the closest the Football League has to a Midtjylland is Brentford. Benham has allowed the London club to operate on similar lines to his other team and I must say that last season’s Brentford side was the best I’ve seen at that club in my lifetime.

However, although they finished the campaign like a train, Brentford have not been as successful this time around and I’m sure many of their fans are convinced that the decision to let Mark Warburton go this time last year was a retrograde step.

Warburton has been described as Brentford’s best ever manager and much of that reputation was gained because of his success in the transfer market. He was reckoned to have an eye for spotting talent and his subsequent success at Rangers, despite spending an amount which I have seen described as “minimal” by that club’s standards, only tends to back up this view.

So, was Brentford’s fine season last year more down to Warburton than the stats which the club set such store by? In this article, Warburton argues that, while he accepts that the use of statistical data is a rapidly growing aspect of football in the 2010s, it should be used in conjunction with the more traditional methods of acquiring players, not at the expense of them.

This exactly mirrors my feelings on this subject. I’d read about Midtjylland at the time of the Manchester United tie and thought to myself that all of this data analysis is all well and good, but it doesn’t tell you about a player’s character and temperament (surely as important as talent when considering whether to invest in him?).

However, the article on Midtjylland I posted a link to earlier shows that the club have a radical answer to that problem – they believe sending someone to watch a potential recruit of theirs in action could be counter productive and so they use scouts to investigate whether they are “a fit from a personal, psychological point of view.”!

Cards on the table, I find the Midtjylland approach (they also pay particular attention to dead ball attacking ploys) to be a bit cold and lacking in spontaneity, but I can’t help thinking that it’s one more and more clubs are going to follow, so are Cardiff City intending to and would they be right to do so?

To try to answer the second part first, I think they might well be right to do so and I would argue that we’ve not been a club where all of the suggestions for new players come from the manager for some time now. For example, in the summer of 2005 (Dave Jones’ first year at the club) we bought in Glenn Loovens on loan from Feyenoord and it was widely reported that we did so because of Peter Ridsdale’s “contacts” within the game.

Similarly, the impetus for the signings of Steve McPhail and Michael Chopra a year later must, surely, have come from Ridsdale who knew the former from his Leeds days and had both players at Barnsley while he was at that club. Besides that, we signed Kevin McNaughton. a player we were supposed to have been after in 2001, that year, so it may well have been that Super Kev was recommended by, say, Sam Hammam.

Malky Mackay might have been interested in Etien Velikonja when he was managing Watford, but he has hinted that he thought he could do better than him once he came to Cardiff because of the increase in spending power he was given. When you consider the way Velikonja was used (or not used to be more accurate) by Mackay at City, the rumours that the Slovenian was a “Vincent Tan signing” look like they may not be too wide of the mark.

It seems to me that much the same could be said about Javi Guerra and Juan Cala with the first named barely used by Ole and the latter only included as a last resort, while Guido Burgstaller (who could well be in Austria’s squad for the Euros and is, reportedly, attracting interest from Southampton) is another one who was seriously under used by City.

Leaving aside any opinion on whether the decision to virtually ignore these players was right or wrong, it does make you wonder nevertheless if the manager’s concerned might have been making a stand against the bringing in of players they didn’t want.

Could it be that an owner who, trying to put this diplomatically, has original and unusual ideas on how the game should be played, has been penalised when it comes to using footballers he played a significant part in bringing to Cardiff simply because the men in question were regarded as “his” players?

Maybe, as I often am, I’m wrong there, but, if I’m not, then the use of a trusted statistical model which is accepted by all concerned (including the new Head Coach) would, hopefully, see a degree of unanimity among the club’s transfer committee and would lead to less of the looking down on the football opinions of the money men by the professionals which I’m sure goes on at most clubs – and has done for decades.

Indeed, when you consider the case of Lex Immers, I have to wonder if something along the lines of the Midtjylland approach has been used already at Cardiff?

I apologise now to those regular readers who have become pretty familiar with this piece in recent months (this is the third time I’ve posted a link to it on here!), but I find it fascinating how much at odds the opinions expressed in the article are from the ones to be found so commonly on all forms of social and broadcast media about the player who made a big impact at City in the last three months of the season.

Just how much you can be influenced by what you read or hear about someone you’ve never seen play before becomes apparent when I remember thinking “we’ve signed a right donkey here” when the Immers loan was confirmed – I always try very hard not to prejudge the players we sign, but, being honest, I did not do so with Immers as I reasoned all of these people cannot be wrong.

Tim Sparv, I didn't want him at Cardiff City ten years ago, but he's Midtjylland's

Tim Sparv, I didn’t want him at Cardiff City ten years ago when I was their manager, but he’s Midtjylland’s “no stats all star” and an essential member of their team according to Rasmus Ankersen.

Instead though, we ended up with a player I recognise now from the description of him in that article. For example, I must have mumbled “good ball” six or seven times on Saturday before realising that it was Immers playing another of those “expected assists” mentioned in the piece on him – it wasn’t his fault that his team mates were unable to make anything of the openings his clever passing created.

Perhaps the decision makers at the club ignored all of the criticism and ridicule about Immers and looked instead at the stats when making a decision on him – if they did, then well done to them, because this is a case where the figures were right and the people were wrong.

Also, when I look at another of our January arrivals, Kenneth Zohore, I see a player who would probably tick a lot of the statistical boxes that a target man type striker for this division might have, but the inconsistencies revealed during his appearances for us (e.g. very good at Burnley and Brentford, pretty good against Bolton and off the pace for much of the time in the rest of his home appearances) suggest to me that the club need to do some Midtjylland style scouting on this player before they commit to bringing him here.

Of course, with Zohore being a K.V. Kortrijk player currently, it’s not certain that he will end up here, but the cooperation between the two clubs in getting him to Cardiff, despite our embargo, is suggestive of there being more to the relationship between them than the fact that they are owned by the same man.

With Ken Choo, reportedly, saying that the nationality of the new Head Coach will not be important as long if they are the right man for the job, the idea that we may be looking at a someone moving from one Vincent Tan owned club to another begins to look more likely to me by the day.

Finally, I suppose the most important question about this stats based approach is does it work? At the time of that piece on Midtjylland, with them on the way to winning their league and Brentford heading towards the Play Offs, those who advocate it would have assuredly said yes. However, with Brentford having dropped four places from last year while never really threatening a Play Off challenge and the Danes currently down in fifth in their league, thirteen points off the top, it now looks like Mark Warburton might be right  - I still wouldn’t be surprised to learn that City intend to operate with a system that is something like the Midtjylland one though.

 

 

 

 

 

 

 

This entry was posted in Down in the dugout, Out on the pitch, Up in the Boardroom and tagged , , , , . Bookmark the permalink.

6 Responses to Is there a “stats based approach” already in place at Cardiff City?

  1. MIKE HOPE says:

    I read this piece immediately after posting a belated response to your end of season blog on the Birmingham game.I mentioned inter alia that the new manager we were seeking sounded like a pen picture of the Oxford United manager Michael Appleton .
    This was largely based on a lengthy and complimentary article in the Guardian about six weeks ago [sorry I can't give a reference ].
    The article was built around the fact that he was regarded as a star of the future until he had the incredible misfortune [or was it bad judgement ] to find himself working successively for the owners of Portsmouth, Blackpool and Blackburn.
    My recollection of the article was that Appleton was a believer in the use of statistical data but it did not sound anything like the mathematical nonsense we were hearing from Brentford’s owners when Warburton was on the way out.
    I was under the impression,perhaps wrongly, that Brentford had abandoned this idea after sacking their first new manager and replacing him with the ex Walsall boss – with much improved results. Does anyone now think that stats alone are the be all and end all?
    Incidentally it was interesting to hear the current Exeter player on this week’s Channel 5 football show being so adamant that Oxford were the fittest and best side in the division.

  2. The other Bob Wilson says:

    Don’t know if this is the article you meant Mike;-
    http://www.theguardian.com/football/2016/jan/29/oxford-united-fa-cup-blackburn-rovers-swansea
    but, after reading it, I reckon there’s a chance that Appleton (we could do a lot worse than go for him in my opinion) might prefer to stay at Oxford after their promotion if we offered him the Head Coach role – incidentally, it would not surprise me to see a struggling Championship club try to employ Russell Slade as manager next season as you suggest in your other post.
    Regarding Brentford, their analytics based approach was still in place in March of this year judging by this article.-
    http://www.cityam.com/237310/rasmus-ankersen-interview-brentford-co-director-of-football-denies-its-all-data-and-robots-as-club-continues-to-utilise-mathematical-modelling
    although Ankersen seems keen to emphasise that there is still a place for more traditional methods at the club.
    I must admit to being fascinated, but not convinced yet, by the analytics approach. I’ve been looking into it a bit more this week and I don’t see it going away any time soon.
    One last thing, it wouldn’t surprise me to see Russell Slade popping up now and again in the Balkans or Belgium, but, equally, as I mention in my piece, I wouldn’t be surprised at all if the movement between clubs is a two way thing with our new Head Coach having been employed by one of Vincent Tan’s clubs in continental Europe.

  3. Richard Holt says:

    A thoughtful and interesting piece Paul. I just wonder whether there is any coherent approach from the powers that be – either stats based or anything else. I confess that my feelings about Slade as manager have been a bit like our weather – changeable. Many times I’ve bemoaned our sterile play, negative approaches, baffling substitutions and such a lack of trust or opportunity for youngsters. Yet, there have always been the mitigating circumstances of the mess he inherited from Solskaer, the total dis-functionality of the club when he arrived (which may have improved to some extent in recent months) and a job description which not many would have been prepared to take on. This time last year I would probably have been pleasantly surprised by an eighth place finish this time around and when we compare our last two seasons with those of say Fulham who were relegated with us, spent heavily at the start of last season and have hardly been out of the bottom third of the division for the whole of that time, it’s not difficult to make a fairly positive appraisal of Slade’s time here. The crux is obviously who comes next. I’d like to see signs of some kind of strategy emerging with this appointment but I’m not holding my breath.

  4. Anthony O'Brien says:

    In the 1980s and ’90s, when the teaching unions began to become a bit more “bolshie”, governments responded with what have become “stats and SATs”. This was clearly not designed to improve educational standards — private schools were allowed much more freedom — but to frighten teachers into doing as they were told. They had to follow a rigid curriculum, teach in a certain way, and submit to the inquisition of a plethora of official “inspectors” – OR ELSE! Sadly, assessment of teachers became a box-ticking exercise of “targets” which could then be summarised by statistics. Only the targets mattered, and teaching was therefore channelled into reaching those “targets”, and to hell with the traditional qualities of a liberal education which withered on the vine. Statistics enabled the “inspectors” to judge a school simply by statistics. Education is like “painting by numbers” and has become a political football rolling downhill ever since. Mere statistics will not improve education.

    By the same token, mere statistics will not improve football standards. They do give the weakest manager, however, the ammunition to keep players in line, to make them do as they are told (think Noone, for instance). Object and you’re out. I can imagine a modern statistics based manager saying to someone like Jimmy Greaves, “I know you scored a hat trick today, but the stats showed you only ran 800 metres and other players did 8,000. Sorry, I have to drop you from the team.”

    I’m not saying that statistics don’t have a place in football (or in education). But what I am saying is that they are not the be-all and end-all. Cardiff’s possession statistics illustrate the point in that they did not tell the story of how badly the team performed since they were artificially boosted by the fact that Cardiff played so much slow tip-tap football in their own half, whereas the better teams moved the ball more quickly and went forward at pace.

    A manager has to be strong enough to back his own judgement and not use statistics as a crutch, only as a piece of the jigsaw making up his decisions. In the words usually attributed (wrongly) to Disraeli, thanks to Mark Twain, there are “Lies, damned lies, and statistics”. Let’s get a manager who can use statistics, by all means, but one who can literally think outside the box! Do I hear Mr Bellamy answering the call?

  5. The other Bob Wilson says:

    Thank you both for your replies. AMO, I think you make some very valid points about the use of stats in general – as for our possession stats, again I see where you are coming from, but the truth is that, we have usually had less than half of the ball in the last two seasons.
    Where I differ from you is that I feel the impetus for this new stats based approach (if there really is to be one at the club) is more likely to have come from the off field side of things than the on field one. I presume that the current transfer committee system will still be in place under the new system and, depending on whether Vincent Tan gets involved or not, the off field contingent will either be the same size as the on field one or in the majority. The fact we have a Committee for this sort of thing shows that the financial management side at the club want to get involved in footballing processes and, if a means could be found where data which both sides trusted could be discussed on a more equal footing, then it could lead to a situation where player recommendations from the financial side might be treated with more respect than they may be currently.
    Speaking as someone who can remember the Brian Clough/Peter Taylor partnership in action at Derby and Forest, it has always struck me how much the former relied on the latter’s ability to spot players who could fit into the system best suited for those teams – would players like Frank Clark, Frank Gray, David Needham, Ian Bowyer, John McGovern and Martin O’Neill have got a look in at Liverpool in the late 70s? Yet. they were regulars in a team which, largely, got the better of the Merseysiders during that time.
    It’s very noticeable how the great periods of Brian Clough’s time as a manager coincided with the times when he worked in conjunction with Taylor – his record at Forest after Taylor left was pretty good, but he’d never have enjoyed the reputation he does now if he did not have the Taylor years to fall back on.
    To be fair, I don’t think the approach at clubs like Midtjylland and Brentford is to do away with traditional scouting completely, but I must admit I found it hard to argue too much with the views expressed in this piece (from the part entitled “Scouting biases” downwards in particular)
    http://statsbomb.com/2016/05/the-death-of-traditional-scouting/
    about the role of the scout today – there aren’t too many Peter Taylor’s out there I’m afraid.
    Richard, I also compare us to Fulham quite often – they are a team capable of playing in a far more exciting way than we do on their day, but they’re so brittle defensively it’s unbelievable and the impression you get from that club, on and off the pitch, is that they are not learning from past mistakes. On the other hand, the efforts of Ken Choo in particular offer evidence that we are learning off the pitch, while, although I don’t like using the word too much, we have “stabilised” on it and Russell Slade has to be given credit for that – it’s not exciting and I’m someone who wanted to see more on so many fronts over the past two seasons, but I stick by what I said in an earlier piece that Russell Slade’s time as manager will be seen in a better light by City fans in ten years time than it is now.

  6. Anthony O'Brien says:

    The biggest problem any new manager with any team has to face from the players or fans is in respect of RESPECT. What level did you play at? What’s in your trophy cabinet? What experience do you have – or failing that, what potential do you have?What exposure did the media give you? (I don’t mean in the case of managers known for their marital problems- but even those issues could be overcome in the case of exceptional players – say, Ryan Giggs or John Terry, or – be it whispered – my own favourite to take over at Cardiff, though less so for the likes of, for example, Brendan Rogers. Respect has to be earned, of course, and as a rule only very unique personalities, such as Wenger , can achieve it without a genuine professional career playing the game. A career cut short by injury is acceptable (think Brian Clough or Eddie Howe).
    Neither Cardiff fans nor the players will react favourably to a virtual unknown, no matter how well produced is his CV. With respect, perhaps that is where Mr Slade faced problems from the very outset. The new manager has to have universal respect from the very start. Surely the door is still open for Mr Bellamy in spite of his tattoos?

Comments are closed.