A Double Dose Of Soccer Analytics

Southern Soccer Scene publisher Ray Alley gives me about 1000 words each month to present sport science topics. If I expand my definition of sport science research beyond physiology and medicine to ‘the match’, a whole new world opens up, including that popularized by the book and movie Moneyball: the field of Sport Analytics.

Analytics has been around many professions for a long time. Finance, retail, health care all have a long history with analytics, but the advent of BIG DATA, where enormous datasets are analyzed has only recently made an impact in sport, including soccer.

Most teams in the top league of a country have an analytics group. What they do with the data they get is anyone’s guess. Big, big, big money decisions hang in the balance.

Now that NBC has gone wall to wall with the EPL, the match watching public will start seeing more data presented on the screen to help with your viewing enjoyment – possession, distance covered, etc. What will inevitably happen is that some long held assumptions and biases will likely be challenged, for fan and fantasy manager alike.

Two recent books have come out and deserve your attention, if you want to learn more about the game. Player, parent, coach, spectator, doesn’t matter. There is plenty here to peruse.

One book is The Numbers Game: Why Everything You Know About Soccer Is Wrong by Chris Anderson and Davie Salley. These[show_disconnected][show_to accesslevel=’Subscriber’] are the guys behind the SoccerQuantified.com blog that I’ve mentioned in two previous columns. The other is Moving the Goalposts by Rob Jovanovic.

Both look at old and new concepts about the game and offer numbers to support or refute various topics. While they are both conceptually similar, they really are entirely different in their approach and presentation.

Anderson and Salley are both business profs, Anderson at Cornell, and Salley at Dartmouth. I couldn’t find an ‘about the author’ page for Jovanovic. You should quickly realize that The Numbers Game is probably a more in depth look into the game.

Jovanovic’s book is a series of short lessons on assumed truths about soccer and most of his applications are from the EPL or the Euros. His number crunching is more of the back-of-the-envelope type. Much of the data used come from match box scores and season summaries.

For example, the myth of the corner. The overall success rate of a corner in the EPL is about 1.5%. In 2008/9, there were 4,232 corners and only 61 goals. A traditional corner is practically giving the ball to the opposition.

How about the assumption that a 2-0 lead is the hardest to hold? Were you aware that in the Euros, the team with a 2-0 lead wins 94% of the time and loses only 0.3%? 2-0 sounds pretty safe to me. Or who was the greatest player? Jovanovic says Pele. Why? Because when Pele played for Brazil, they had a 94.2% winning percentage. When he didn’t play, Brazil had a 56.3% winning percentage; Quite a dropoff.

But what about Maradona? Surely he was the greatest. Sorry. When Maradona played, Argentina won 62.2%, but were actually a little better and won 69.4% when he wasn’t in the lineup. OK, so maybe ‘greatest’ isn’t the best term, maybe ‘important’ is better in this analysis.

Anderson/Salley offer a far more in-depth look at a few major concepts of the game and in doing so, use bigger datasets like those from Prozone, Opta, databases available just to the media, and even the EPL’s accountancy firm. BIG DATA by comparison.

To begin, they discuss in some detail the role of luck in the game and offer quantitative explanations for why the outcome of a game is almost 50% pure luck. Tied to this is the fact that the infrequency of goals makes luck an even bigger role than in games with more scoring like basketball or American football.

They also tackle (no pun intended) the concept of possession vs. direct play pointing out while there is little correlation between time of possession and outcome, teams that are able to possess the ball not only score more, but they also prevent the other team from scoring.

And they also provide some insights into teams that buck this trend like Stoke, Watford, and others. Like how Stoke can have less than 30% of the possession, take 2 shots, and beat Chelsea 2-0 last year.

Here’s a question bound to raise the ire of some. Which is better? To score or to prevent the other team from scoring? Think about that. You can score a goal and still lose, but if you shut the other team out, the worst you can do is tie.

And from this, they go into a discussion about formations (is catenaccio really a defensive formation?), personnel, managers, management, and decisions about transfer and fees being so high for strikers and so low for defenders. Their position is that defenders are ridiculously undervalued. As a former defender, I knew that!

One final question. Which would you rather your team be? Eight (outfield) players who’d be considered as a 7-8 out of 10 and 2 stars who are 10 out of 10 or a team of 10 outfield players, all 8 out of 10?

This goes back to their comments about team makeup and overpaying of stars (i.e., strikers). According to their calculations, it’s the weakest link on a team that defines outcome, not the superstar.

So while both books are about ‘analytics’, they each have a bit of a different slant on the topic. Jovanovic takes more of a ‘People Magazine’ approach to analytics. Short, easily read and digested reports. The perfect presentation for those who want to have some facts in their arsenal for inevitable debates.

Anderson and Salley provide a deeper journey into broader concepts of the game (while sparing the reader from the gory mathematical details).

It all depends on what one wants out of their ‘analytics’ so I won’t recommend one over the other. Each will set you back $12 to $16 depending on where you find it. At that price, get both.