Jump to content

 

 

Moneyball - Can statistical scouting be used by Rangers?


Recommended Posts

Enjoyed this though-provoking submission from @JohnMcIntosh on our main site:

 

Article is reproduced below but please take a moment to click the link!

 

http://www.gersnet.co.uk/index.php/news-category/match-analysis/401-moneyball-can-statistical-scouting-be-used-by-rangers

 

What is Moneyball? Essentially, it is the process of finding true value in the transfer market through others in the game undervaluing them via statistical analytics and set variables. There are set rules to work to, such as “use the wisdom of the crowds” and “sell any player who is offered their transfer value and replace them before they leave”.

 

Of course, the practice has become more popular after the film “Moneyball”, starring Brad Pitt as Oakland A’s General Manager Billy Beane, was released. The film showed the genius of statistical analysis, finding value in the market through players being undervalued by other clubs, as well as the reluctance of people involved in the game to embrace this new found “nerdy” way of looking at the game of baseball.

 

In many ways, the same can be said of football. Yes, statistical analytics in football is way behind the major American sports and, although it’s use is growing, the sport will almost inevitably never get to the same levels of reliance on analysis as baseball has, due to the fluidity of football compared to Baseball. And for me that's fine, but it is interesting to look at how it can be used across wide-ranging topics from opposition tactics, player recruitment and comparing your own team to your rivals or the rest of the league, to examining a declining player’s statistics to identify exactly when the time is right to move them on.

 

Arsene Wenger is a massive believer in statistics and near the end of Dennis Bergkamp’s career, he showed him just how his performances had started to deteriorate in the last 30 minutes of games. Using this method, Wenger has been able to, in most cases, sell his top players at the age of 28-30 for good prices - and very rarely have they ever went on to perform at the same level once they have left and I find that fascinating.

 

If we compare the Oakland A's roles to their football counterparts, it would mean taking the General Manager position and replacing it with a Director of Football, or as it is more commonly known in Europe, a Sporting Director. For the decision making process, it is very similar with a “transfer committee” in operation (that word probably makes you shriek and think of Liverpool, but it is common and successful elsewhere).

 

It is very common in Europe for teams to use this strategy, with varying levels of reliance on data analysis to form decisions on player recruitment. In some cases, the “Head Coach” is excluded from these discussions, and although I can see the merits for doing this, in reality it can cause a lot of friction between the Sporting Director and Head Coach in a relationship where it is so important that there is mutual respect and clear boundaries.

 

There is a great part in the excellent “Soccernomics” book, by Simon Kuper and Stefan Szymanski, where they discuss the moment when data analysts came to then Manchester City Manager Roberto Mancini, stating that more goals are scored from in-swinging corners to the front post, though the goals tend to be more scrappy, and Roberto dismissing this as nonsense and saying that out-swinging corners would continue to be employed by the club – beautiful goals are scored from out-swinging corners so our brain remembers them more, but the stats don't lie.

 

A Director of Football now seems very unlikely at Rangers, with Dave King stating it is not necessary at this time and being quoted as not fully understanding it. Couple this with his comments on us needing more of a coach than a manager, to nurture and develop talent, and it doesn’t quite match. These two statements for me clash and a Sporting Director looking long-term with a Head Coach focused on the playing squad and the short-term would make perfect sense.

 

We have come a long way in recent years, with the two major statistical analysis companies involved in football, Opta and Prozone, offering their considerable services to clubs. How far their service goes depends on the club’s willingness and how many analysts the club has internally. Both of these companies can send someone out to break down the team’s statistics and show what these mean in practice. Or, if there is a data analytics team within the club, they can just provide the statistics.

 

If we look at Scottish football, we are many, many years behind even other major footballing countries. To my knowledge, one of the only clubs in Scotland who do focus on this are Hearts, though clubs tend to be secretive about this so it is hard to tell.

 

So what statistics can be looked at to find a certain player? Let's say you want a central midfielder who is a ball player, so you search for pass completion rates of over 80% and someone who has played over 20 games in a specific league in the previous season and the data shows a list of players who achieved that. One statistic alone is not enough though, and you would need to view other related statistics for a player you like the look of, delving much further and ALWAYS going out and scouting the player whilst getting various scouts to look into that player (employing the “using the wisdom of the crowds” rule), data analysis can only assist a football team, in my opinion: it can't be the decision maker.

 

Also, when you search for, say, over 80% pass completion in say the premier league it will show up the usual big hitters – Gerrard, Silva, Yaya Toure etc. BUT, strangely enough, it would also have flagged up Kevin Nolan a few years ago which is something you would need to look further into but is interesting nonetheless.

 

But there has to be more to it than pass completion? Yes, of course there is. Statistics used range from how far a player has run in a game, breaking this down to see how tired players get in last 30 minutes, tackles made, assists, speed of assists/goals once crossing half way line, high intensity output, goal ratio within the 12 yard box to the more usual goals-to-game ratios or, if you are looking for a “Claude Makelele,” you could check a heat map for positional awareness and how many interceptions they make alongside high intensity output to form a judgement - but there is far more statistical data for forward thinking players.

 

If we take a look at the major football clubs to use this “moneyball” way of working, Lyon perfected it between 2002 and 2008 and were able to corner the transfer market using data analysis, signing players for very cheap, nurturing and developing their talents then selling them on for massive transfer fees, by which time they had replaced them already. Amazingly, in this period they won 7 French Ligue 1 titles in a row and had 4 different “Head Coaches”. Their background staff didn't change too much in this period, so the Club always had stability in that sense.

 

The best clubs you will find these days who use statistical analysis are Sevilla and Southampton, both of whom employ a Head Coach/Sporting Director dynamic and both of whom manage to gain large transfer fees whilst replacing them with very cheap deals. Sevilla’s success is slightly more impressive given they don't have the benefit of the inflated transfer values found in the English Premiership.

 

For example, Sevilla just sold Aleix Vidal to Barcelona for £17 million, having bought him for £2.6 million. Vidal follows in the footsteps of former Sevilla player Dani Alves, bought for £450,000 and sold to the Catalan club for £32 million. It is a testament to their Sporting Director, as well as their impressive Head Coach Unai Emery, that they can continue to find this type of talent, know when the right time to sell is and are still capable of putting out a consistently competitive side that can win things - Sevilla have won the Europa League 4 times in the last ten years.

 

Closer to home, Brentford signed our wee Lewis in January. With their owner a massive advocate of statistical analysis, I would be massively interested to know 1) how big a part statistical analysis played, or was it more to do with conventional scouting from the start, and 2) what set of statistical variables alerted Brentford to him.

 

In conclusion, I would love Rangers to hire a team of data analysts (very cheap to get) to look into the data analytics side of things far more, it can be used in a wide range of topics and ultimately massively benefit our club. Paul Murray has said that we will be a modern, forward-thinking football club and this is one part of that vision that I feel needs to be addressed.

Link to post
Share on other sites

It's another modern practice that should be implemented at Rangers; one of many I might add. We need to modernise, and perhaps find a niche, because there is no way we can compete with most clubs that should our direct rivals (not just in Britain, but Europe also).

Link to post
Share on other sites

The thing to be careful about is that there is a massive difference between baseball and football. Baseball is all set pieces and a lot of what players do can be isolated and a "golden" metric found to measure their effectiveness. A first base man will pretty much do the same thing, no matter who else is in the team. Football is a far more fluid game where everyone can interfere with everyone else and is more reliant on how each team plays together as well as how they cope with the other team. A golden metric for each position could be pretty hard to work out.

 

The threads on Boyd are an obvious example of this. Players who play badly for a team and go on to play better for another, or vice versa, are other examples that don't tend to be so stark in baseball - therefore if you trade a player for the Mets from the Red Sox then it's likely his statistics won't change so much. No so much in football. Your stats could be impeccable but the team you build pretty crap. I would be confident in asserting that doing it all statistically like this, our squad would have been predicted to win the Championship.

Edited by calscot
Link to post
Share on other sites

PS Moneyball showed that the advantage gained comes from being the only team to do it, if you show it works and everyone starts doing it, the richest teams win again. If we open that pandoras box and it works, we'll soon be back to square one but then find it even harder to compete with the TV rich teams.

Link to post
Share on other sites

I don't know enough about baseball to comment but I do feel there's a place for statistics in football when it comes to our footballing and scouting operations.

 

Trends are trends and well worth exploring but only if it's done correctly. If not, it may just cause more harm than good.

 

Maybe just a Chief Scout of repute would be a good place to start... :(

Link to post
Share on other sites

 

Maybe just a Chief Scout of repute would be a good place to start... :(

 

Open to abuse here, but I'd take another year or 2 of re-building/Championship if it meant spending money on these type of guys, and building a philosophy at the club akin to Ajax/Barca/Anderlecht etc. Are we talking about a Cathro/Warburton type of manager here?!

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.


×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.