ian-wright-snapshot
Statistical Data Analysis

Could a computer have discovered football legend Ian Wright?

Led by the moves of Juan Cuadrado to Chelsea for £26.1m ($40.3m) and Wilfried Bony to Manchester City for £25m ($39bn), England’s Premier League alone accounted for £130m ($200m) in football transfers in the January transfer window when deal-making reaches fever pitch. Accountants Deloitte said a record £950m ($1.46bn) had been spent over the 2014/15 Premier season. The stakes are high but for every player that carries an eight-figure price tag there are thousands of players that don’t. Finding that one jewel, that one star is what every scout dreams of whether it’s football, rugby, basketball, baseball, rowing or NFL football - but how do you do it when the odds are stacked against that hope?

In 1995, I interviewed the late Peter Prentice, chief scout at London’s Crystal Palace FC at the time, and the man that ‘discovered’ Ian Wright, the future Palace, Arsenal and England striker. Prentice told me that you get an instinct about certain players. A humble and very likeable man put scouting down to just that – a human feeling – and he believed it was impossible to work to a formula, due to the varying nature of players, personalities and skills.

Yet today, sports clubs around the world are doing just that: using algorithms to analyse potential players by mining data on performance, fitness but also personal habits and attitude. Most, if not all, top professional clubs now have a data performance analyst as part of their setup, but not all use data to assess the potential of new recruits.

But Ian Wright was nearly 22 when he signed his first professional contract in 1985, which goes against all the usual rules of development and talent discovery that you have to “get ‘em when they’re young.” Ask any scout and they will usually say that no-one slips through the net, but of course they do. So is this why data analysis is winning over youth development officers across the world?

Brian Prestidge is Head of Analytical Development for the historic English football club Bolton Wanderers FC and believes that applying analytics in a creative and dynamic environment such as football is “a challenging concept”, but one which many clubs are now taking very seriously. Bolton uses a US data analytics system called Tableau.

“Traditional analysis in football has been video-based with some basic metrics,” he says. “However, in the last few years this has changed drastically with clubs setting up strategic processes utilising data analytics as the driving force,” he says. “Analytics in succession planning provides not only an objective means of assessment and future planning, but also a way in which to predict player development and future performance for various benefits: minimising mistakes, contract offers, and development plans.”

Prestidge doesn’t go so far as to suggest that the system could discover a late developer such as Wright but the inference is that a data-based approach to analyse potential players should reduce errors of human judgement and enhance decision making. The sporting and financial gains of making accurate calls on potential playing staff are not to be underestimated.

With the implementation of financial fair-play rules that control spending on transfer fees and the desire from the English FA to have more native players competing in the top leagues, “clubs are now exploring every avenue in an attempt to determine not just a player's potential, but by employing predictive methods, the probability of a player reaching his potential”, adds Prestidge.

 

One and done

Data analytics is not exclusive to football either. Other sports have also been plugging into data to discover and assess young talent for some time. In the US, Tibco CEO Vivek Ranadivé has been taking his own medicine, using the company’s data analytics software for his own NBA basketball franchise, Sacramento Kings.

“I use Big Data for everything,” he says. “On the game side, I looked at 30 gigabytes of data.”

He talks about the problem of recruiting rookies and not knowing if the big contract is going to pay off because some rookies are great for a season and then that’s it.

“They call it ‘one and done’,” adds Ranadivé. “So you need to figure out whether this 19-year-old kid is going to be a player for the long term. I did a combination of crowdsourcing with machine learning and neural networks software, to help me predict this kid’s future – what NBA player might he look like in three or four years and what is the floor and ceiling on that? Based on the results I picked my player. I am using tools that have never been used before to do this, so we’ll see how it works out next year.”

While Ranadivé has a bit of a waiting game to test his theories and formulae, other sports have been playing with data for some time. British Rowing for example now has a well-developed scheme called the Start Programme, to capture data on a regional basis. It measures rowers (or even non-rowers) for their potential in the sport, putting them through their paces on a variety of strength and endurance machines at training days. The resulting data is crunched and matched up against data from elite members of the GB Rowing Team to assess potential.

Laurie Miles, Head of Analytics at SAS UK & Ireland, helps to support the British Rowing Start Programme and claims that “by collecting and analysing data on young rowers as well as members of the elite GB Rowing Team, British Rowing can build up a clearer picture as to whether a young athlete will flourish in a couple of years and be eligible for youth international representation.”

Gamification

The successes speak for themselves. At the London 2012 Olympics rowers that had been discovered through the Start Programme won five golds and one bronze and made up nearly a third of the entire team. It is this kind of success that all other sports teams are trying to emulate, none more so than football. 

In August last year sports data company Prozone decided to include stats from popular computer game Championship Manager in its new service Prozone Recruiter, to help scouts analyse and identify the stars of tomorrow. What? A computer game?

It’s not as far-fetched as it sounds. For years Championship Manager has gleaned player data from scouts and fans across the globe, putting numbers and growth potential to players on a range of characteristics and abilities from pace and shooting through to temperament and injury history. Plugging this data (which, unofficially, scouts have been using) into an official recruitment tool makes perfect sense.

In a statement, Prozone CEO Thomas Schmider said that Prozone Recruiter “has been built to supplement the intuition of scouts and coaches by delivering detailed performance information on over 80,000 players worldwide.”

Never before have scouts had so much information at their fingertips. As clubs increasingly use data as part of youth development and analysis and have tools such as Prozone Recruiter to plug into, the idea that anyone could slip through the net surely diminishes. Clearly, the life of a scout is changing dramatically but with increased data analytics tools will come increased pressures to succeed. What this shouldn’t do is impair the scout’s intuition and lead to an over reliance on Big Data.

Prestidge at Bolton says that “the focus of data analytics must be on controlling the ‘controllables’”, especially as there are so many ‘uncontrollables’. Tools should remain just that, he adds.

So could Championship Manager have discovered Ian Wright? Anyone who has played the game and been sucked in by its addictiveness over the years will know the answer to that. For those that haven’t, the answer is quite simply, ‘yes’.

 

Also read:

No more cold burgers as teams target sports fans via data analytics

Can tech lead to football fever in India?

PREVIOUS ARTICLE

« Eye-tracking in games: The next analogue stick?

NEXT ARTICLE

FCC approves net neutrality rules »
Marc Ambasna-Jones

Marc Ambasna-Jones is a UK-based freelance writer and media consultant and has been writing about business and technology since 1989.

  • Mail

Recommended for You

International Women's Day: We've come a long way, but there's still an awfully long way to go

Charlotte Trueman takes a diverse look at today’s tech landscape.

Trump's trade war and the FANG bubble: Good news for Latin America?

Lewis Page gets down to business across global tech

20 Red-Hot, Pre-IPO companies to watch in 2019 B2B tech - Part 1

Martin Veitch's inside track on today’s tech trends

Poll

Do you think your smartphone is making you a workaholic?