As the world continues to learn to adapt to a new normal, data has never been more prevalent, more important, or more valuable to our everyday lives. It informs us of everything we do, have done and are likely to do. It keeps us safe, enables us to buy whatever we need and creates untold sources and avenues of conversation, entertainment and engagement, all at the touch of a screen.
Crucially, it allows us to dive deeper into previously unpredictable events and chart a clearer picture of what lies ahead. At Sportradar, we are pushing the boundaries of data intelligence to rapidly transform the company from the world’s leading betting data provider to a leading global technology business. By using new data-driven technologies and methodologies, we are empowering our partners and their customers with instant, real-time visions of the future, and have identified four key areas that will shape the future of data in 2021 and beyond:
AI and machine learning technology
I have always held the view that data is both the source and engine of knowledge, and for the last 20 years I have based our own business model on this philosophy. Today, however, the sheer volume of data we have at our disposal requires new technologies and new ways to collect, process and distribute this information to customers and consumers at lightning speed, with total accuracy and minimal to zero latency.
Artificial intelligence has been with us for some time, but we are now beginning to see its growing influence, which will skyrocket over the next few years with an increasing number of businesses using AI to drive operational efficiencies.
As an advanced technology business, we have long been believers in the future value of AI, but its impact is now becoming clearer. From instantly modelling new sports, sporting formats and future events weeks ahead of schedule to the rapid creation of new data-driven digital betting products like our own AI-driven and algorithm-based Simulated Reality, which was created and brought to market in just 10 days in March due to the void in live sport, machine learning is transforming the way sports data is being designed, digitized and delivered.
Computer Vision and automation
A core area of machine learning that is rapidly being adopted is Computer Vision (CV)—a form of AI that teaches models to interpret and understand the visual world. This aids the creation and training of data-driven models to anticipate the probability of significant events based on visual patterns of play, enabling increased accuracy in live sports, as well as more precise and far-reaching modelling of future events.
At Sportradar we are stepping into a new technology era with CV radically transforming the volume and speed of data points we collect. Today we focus on game-level events (corner, goal kick, attacking play, etc.), tomorrow we collect and process data on all players on the pitch (where and how they are positioned, down to the position of their individual joints), enabling us to develop new data-driven sports markets and products. This might seem to be over-kill but with this we can derive new data points like player acceleration and type of shot.
Within our data scouting network, for example, we are working on training a sophisticated speech detection model where our data journalists use spoken commands to map every element of a live game from key moments to weather conditions, thereby instantly broadcasting live data into our network and improving latency in betting markets where timing is critical. Speech detection will remove 3-4 seconds of latency in getting data to our customers.
Investing in technological resilience
Consumer tolerance of technical failure has decreased dramatically in recent years and this will only continue. The exponential rise in expectation levels represents a major challenge for any company working within sports data and covering hundreds of thousands of games in real-time with sub-second latency.
There will be an increased shift towards large investments in technological resilience, and transformative measures aimed at fast-forwarding businesses and investing in the latest public cloud infrastructure will be essential. This enables maximum sports data provision to deliver enhanced scalability, durability, latency and elasticity.
At Sportradar our aim is to always be at the leading edge of technology. We have therefore invested in what we call our “Ring of Steel”: a dedicated, global, cloud-based backbone network that enables us to distribute live data anywhere in the world, improving reliability and latency by up to a factor of six, with zero disruption and massively improved security to any customers.
Data for good
A phrase I often refer to is that “facts dispel myths”—the more we can use data to tell not just a story, but an accurate story of what happened on the pitch, the better. This is critical within sports integrity where data is used to detect anomalies and suspicious betting patterns in games. While historically this would have involved data analysts examining data streams in the search for indicative outliers, today it is a combination of machine learning alongside data analysts who can confirm suspicious behaviors found by the algorithms.
In 2019, Sportradar Integrity Services detected 660 matches in global sport as suspected of being fixed for betting purposes. With the COVID-19 pandemic causing significant financial hardship for teams, individuals and leagues, match fixers are certain to prey on them. The combination of data, technology and intelligence will remain vital in the fight against corruption and betting fraud.
But it is not just match fixing and integrity where data is being used for good in sport. Most recently, we launched an industry-first solution to support sports organizations in safeguarding professional athletes from social media abuse. This data-driven service is designed to protect the mental health and wellbeing of professional athletes by keeping them free from harm online through successful investigation, proactive intervention and disruption.
Technology and data always drive innovation and with exciting new technologies like 5G working alongside AI, machine learning and computer vision, the way in which fans consume sport is certain to change and improve. The data we have at our disposal today really is infinite and, if used in the right way in conjunction with new technologies and methodologies, will empower us all to shape a better, brighter future for the sports industry.
Carsten Koerl is Chief Executive Officer of Sportradar AG. Carsten has strategically led all aspects of Sportradar’s business since founding the group in 2001, driving the growth of the company from a tech-focused start-up to a global technology leader in the sports industry. Prior to founding Sportradar, Carsten held a number of management positions within the software development and gaming industry. He was the founder of bwin and led the company through one of the most successful public listings ever undertaken on the Vienna stock market. He holds a Master’s Degree in Electronic and Microprocessor Engineering from the University for Applied Sciences in Konstanz (Germany).