Opta Sports Data began in 1996, initially to create a performance ranking for players and team. It’s now critical to a huge variety of customers including football clubs.
Every touch of the ball is monitored live from over 2,000 games of football a year resulting in over 4,000,000 individual events.
Opta provides the full analysis from the English Premier League, Italy's Serie A, the German Bundesliga, France's Ligue One and the UEFA Champions League.
They provide live score information for the English Championship, League One and League Two, the Scottish Premier League, Bundesliga 2 and the Italian Serie B.
It started in the pub, where the company’s founders watched games on TV and wrote down the events that happened on the pitch using pen and paper. Then they started to use Excel to replace pen and paper and from there it mushroomed. Now it’s a complex software landscape – doing the same thing but more in depth, faster and in real time.
From an engineering point of view the fascination is in using MS framework technologies to collect the data but then to stay on top of the development. The software is based on open source components and if features are missing we supply them ourselves. For example, information needs to be transported so we developed a messaging system. With the strict requirements for timing, email wasn’t good enough so we have messaging that goes out from our databases to the customer.
One challenge was the economic pressures and timeframes. When you start as a small company you might have 2 weeks to produce a bespoke product, so you have to make compromises in the technology. Having taken short cuts to get it out on time, you then come back later to polish it and make it ready to sell as a package.
The information we produce goes out to journalists immediately after the match and they receive it as an xml file.
I’m not particularly into football but what we have at Opta is customers who create fantastic products out of our data – 3D animations for the web, mobile internet graphics so that people can follow the match while they’re on the train.
Information goes to clubs so they can use it for training purposes.
We have a standard package for our media customers, and producing it is the easy part, but then we need to evolve and adjust it slowly.
One product might be pre-producing the content for media companies, so what you see when you visit the Telegraph website football stats pages for example, the look and feel of the page is what we pre-produce ourselves. It’s customer-specific.
A UK premier league game might be announced 6 months before it happens so our data collection (aka Ops) team manages the schedule and as soon as the information comes in they put it into the staff planning tools so that 2-3 people per game analyse the match. There might be 10 games going on at the same time so 20-30 people. Mostly freelancers.
These ‘analysts’ are football specialists and don’t need any technical expertise apart from knowing how to use a keyboard and mouse. They’re specialists in a particular league and good at looking at the game and understanding immediately what’s going on, who they’re seeing (because you don’t always see the player’s number on their shirt while watching on TV). If it’s a UK v Germany game they’ll need training on the German team, as they’re only specialists in the UK leagues.
What they do is key in ‘events’ with the ball – passes, goals etc -while watching the game in our offices. The focus is on accuracy . It’s positional – where, what and who.
Time stamps need to be very precise because one of our customers produces automated video clips, so if a goal event takes place at 12 o’clock they cut their video clip from our data for post-production.
The info goes straight into the database in the data centre in London and modules are triggered by certain events e.g. goals to create content for the customers and this is sent out automatically.
I see a lot going on in the consumer markets – games are almost more popular than the real sports. For example Fifa 2009 has broken sales records. That’s a big opportunity for us to have our data in those products, to power those games.
Technology will be more and more important for decision-makers. For example if they’re searching for a new player, by defining their criteria they can have comparable data to compare for example how a low league team from Holland compares with a low league team from South Africa. They can base their decisions on facts, not gut feelings.
Enter the Faraday 08/09 competition and win!
Play Science Friction (game)