The global sports market is worth somewhere in the region of $520 billion, and growing. That is a lot of money. Where there is that much value in a market, other business sectors follow. And, if you know sports, you’ll know how much data collection has changed the game in recent years.
Teams and sports organizations now collect an incredible amount of data on athletes’ matches. This, you won’t be surprised to know, is rarely kept in-house by those who gather it. Instead it is licensed out for a variety of purposes from internal research and operations, to sports betting odds makers, sports journalists and data providers, sports science research, fantasy leagues and much more. This is the technological backbone that powers this massive economic sector.
A Brief History of Sports Data Collection
Today, sports data is massive international business. A sportsbook like Betway Ghana will have vast amounts of data from American sports leagues, and many other places, informing its betting odds, as well as what kind of teams and players are popular choices for promotional odds boosts.
Going back, the earliest form of sports data collection began in the 1850s. Primarily in cricket and baseball. Henry Chadwick began recording batting averages in box score charts in 1858. However, these methods relied on pen and paper and were largely only used as a retrospective curiosity or for affirming records.
It wasn’t until the 1970s that the modern approach to sports data began to form, with the Society for American Baseball Research and its pioneering founding member Bill James. Kansas-born James was an early advocate of sabermetrics, derived from SABR. He was one of the first people to look at sports data’s propensity for predicting future results, and how that might work.
By the 80s and 90s coaches in all kinds of sports were recording matches on VHS tapes to look back over for performance analysis and opponent scouting.
However, the next step to becoming the massive multi-billion dollar sports data market of today was taken in the early 2000s. The advent of Moneyball, wherein teams with low budgets use computer-driven data analytics to identify undervalued players, changed the game for sports around the world.
How and Where Data is Collected
In the old days, sports data was recorded by someone watching the game and noting down events. Today’s sports data collection barely sees human involvement (beyond the athletes) while the game is in play.
Biometric sensors. Complicated camera systems covering every blade of grass or inch of floor. Sensors built into balls, rackets, helmets. If you can name a data point in sport, someone has invented a way to track it.
Once it is collected, the data points are stored in vast databases – usually where multiple different organizations can access and use them.
Data Goes to Sportsbooks, Sports Scientists and Others
For example, consider NFL data. Zebra Technologies collects the data, using a network of receivers and tags placed around stadiums. During matches this is sent in real time to Amazon Web Services, which employs machine learning models to turn raw spatial data into readable, usable metrics for game broadcasts.
After the game, the data is sent to Genius Sports – the firm in charge of distributing NFL data. All 32 teams are given their own synchronised copy of the data set for operational management including evaluating opponents and assessing player fitness and form. The NFL itself also uses the data set to look at improvements to the game including monitoring player safety and injury risks.
Then it is distributed to the syndicated subsets as per their need via API feeds. These third-parties include sports betting operators, fan analytics providers, sports video game makers and sports media platforms. In some cases even stakeholders like player agents, unions and insurers might request access to the NFL’s central data, although historical data is widely available online.
The Challenges and Risks to Mass Data Collection in Sport
In many ways, mass data collection has made modern sports into the absolutely huge business it is today. Data analysis has empowered teams and individuals to reach the very highest levels of their game, in ways athletes in the early days of pen and paper data collection could scarcely have imagined.
The spread of sports media, including video games, to which accurate data is an integral selling point, has also intensified interest in sports more broadly – as has the global growth of sports betting.
Yet not everyone is happy about the situation. The state of play in modern sports has significant risks around data ownership, security and privacy. Centralized data systems are always somewhat vulnerable to attack, while syndicated systems have the potential for data to spread without consent.
According to one study of professional soccer players in Europe, 78% of 150 players surveyed said they had concerns about potential misuse of their biometric data. This is something the sports data business will have to tackle in the coming years, even as all the sectors around it continue to grow.