Crunching the numbers to land Premier League points

24th March 2017

Hidden away behind a desk at Swansea City's Fairwood training base, Alex Thomas digs into data on every player in a bid to help Paul Clement's team win matches.
Here, he explains why analytics are becoming ever more important in football.




He is perhaps the only member of Swansea City's backroom staff who rarely, if ever, gets any mud on his boots.
"I was given a pair at the start of the season," Alex Thomas says through a smile, "but I haven't been out on the grass yet."
When the Swans are out training, Thomas is in a familiar spot.
Pay a visit to the club's Fairwood training base and, whether it is first thing in the morning or late in the afternoon, the likelihood is that Thomas will be sat crunching numbers at his desk.
Such is the life of a data scientist.
There are not many of them around in the Premier League. In fact, Thomas suggests, what the Swans are doing in this field is on a par with similar work being done at European super clubs such as Bayern Munich and Paris Saint-Germain.



"And we are pushing on," he adds.
"Football is getting more technical. It is data-rich now and the skills needed to work in football are growing every day.
"People want to be the best."
Thomas is never going to be thrust into the limelight.
The hope is, however, that the work he does behind the scenes will help Paul Clement's team grab the headlines.
Thomas's job is to compile and study data on every Swans player - as well as opposing sides and potential signings - so that the right information can be passed on to Clement, his squad and the rest of the backroom team.



"We record everything the players do - how much they sleep, every rep they do in the gym, every pass they make in training. All their physical data and technical data is recorded," he explains.
"The job of data science then is to try to find trends or anomalies - you are looking for the usual things and the unusual things.
"We can look to prevent injuries, for example, if the data tells us a player is lacking in some way. We are getting meaning out of the data."
If it sounds complicated, that is because it is.
Born and raised in Swansea, Thomas did a degree in maths and physics, then a Masters in theoretical particle physics, at the University of Edinburgh.
But it was a stint after university spent travelling the globe as a ski instructor which eventually led him back home.
"That was what sparked my interest in working in sport full time," Thomas says.
He was doing a PhD at Swansea University which involved researching sports data when he was asked to do some work as a consultant for the Welsh Rugby Union.
Next came a similar role at the Swans, and within a couple of months Thomas was offered a full-time job at the Liberty.



"I have been a Swans fan all my life and was a season-ticket holder before I got the job," he says.
"I stopped my studies to come and work for the club and it's really enjoyable. I enjoy problem-solving and I get lots of things thrown at me every day here."
Thomas's skills mean he can help others in the backroom team get their work done.
He has developed the software which allows every staff member to feed various data on players into one programme.
That means information on any player is available in one place for all within the club to see.



"When we first started (during Michael Laudrup's reign) there was no centralised database," he says.
"Everyone was working on their own island. So if one person was taking a player out for a rehab session, they would record the numbers and keep them on their computer and tell someone if they wanted to know. That's how a lot of clubs worked.
"But we have taken a big step forward by putting in place an athlete management system which everyone can access.
"Everyone has a broader picture now. You can get that picture much more quickly about a player."
Thomas's work can also be useful when the club are assessing potential new signings, while he also has a part to play in the Swans' preparations for upcoming opponents.
"We rank all the teams in the league," the 30-year-old says.
"We can see that this team likes to cross the ball a lot, for example, so we can tactically work towards dealing with that.
"I produce the numbers on these teams, but then it is down to the people with the expertise to interpret them."



Thomas says that different managers have differing views when it comes to the significance of analytics in football.
But his task is to sift through the numbers to highlight relevant points, then it is up to the football staff to act on them.
"Think of it as a big iceberg," he adds.
"All of the data is underneath but the meaning is the bit you see on top. There is loads of work underneath, lots of data that has to be worked through. But the interesting bit is at the top for everyone to see."
A modern coach, Clement is keen on all the new tools which can assist his team.
"He knows what he wants," Thomas says.
"When he came in the pre-match report (which is produced by the Swans' analysts) went from five pages to 30.
"He is very structured in his approach and very organised and interested in the facts when you can give them to him."
And so far, it is paying off.