• Blog:

  • Home
  • Ably News
  • Ably Engineering
  • Developers
  • Industry Trends
  • Hardest Aspects of Realtime Engineering
  • The real(time) future of sports data

    By: Eve Martin 6 min read

    Coming from leagues, fans and regulators, demand for realtime sports data in has led to a proliferation of streaming APIs. Are current data transfer systems fit to support a fast-changing industry?

    With  the ‘sports data revolution’ now old news, the next wave of data-driven  transformations is being driven by realtime. A combination of connected  fans (the subject of an Ably/ OnePulse survey into mobile sports consumers), newly opened sports-betting markets (notably the US, predicted to be a $215bn betting industry by 2025), and the rise of ‘continuous’, in-play bet-waging have given rise to an explosion of streaming APIs. What does this mean for data transfer infrastructure in sports?

    To illustrate, Ballr is a mobile gaming platform where fans connect, compete and  communicate, in realtime, around pre-defined in-game ‘events’ (e.g., a  pass, a foul, a change in weather or pitch condition). Engineers must  create and maintain systems that transfer data on thousands of  ‘micro-moments’ — dynamic updates on play, betting odds, user  preferences, and other information. Millions of messages require  thousands of data-transfer channels, all running simultaneously, with  little certainty around which channel or message will be relevant when,  to whom and to how many.

    Infrastructure for fast, fickle, realtime-addicted consumers

    Infrastructure  requirements for realtime streaming APIs come with high-stakes strings  attached. For sports betting, where competition is fierce and users  stand to lose money, data must be delivered to end-users in under half a  second and seamlessly synchronized across users’ multiple devices, or  sports fans (nearly 70% of them, to be precise) simply switch apps at  the first sign of inconvenience. The technical requirements for meeting these heightened realtime expectations are the subject of detailed research carried out by Ably last year into the connected fan phenomenon. While competition is fierce and consumer expectations high,  the benefits of making the most of this opportunity are also  significant. Live data streams from sports events by definition attract  large audiences. Easy exposure to these APIs means more revenue streams,  as partner businesses add to live fan experiences (see Fresh8 example,  below).

    Supporting a competitively innovative industry

    While live sport remains one of the most popular — and lucrative — entertainment channels, TV viewing figures are in decline.  A race is afoot to capture the new, mobile-based audience by delivering  the best, most novel and most immersive fan experiences possible. What  do these look like in practice?

    In-platform betting experiences allow bet-wagers to access data streams on demand and set their own terms for bets. Fantasy Football Fix is  a platform for social betting experiences. The infrastructure behind  this platform is impressive: individual in-game actions are collected,  packaged and distributed while being run through dynamic, AI-driven  predictive algorithms. They are then distributed back — in real  time — to thousands of players. As well as performance, low latency and  scalability, it’s also important these data streams form the basis of  other innovations. Forerunners in leveraging realtime streams on  multiple aspects of a game include Fanduel, which recently partnered with Fresh8 to integrate realtime image- and video-based advertising across its channels.

    Traditional data streams — a blockage?

    Data  is currently streamed through networks that rely on pairwise  integrations. Matthew O’Riordan, CEO of Ably Realtime, explains further:  “Existing data transfer models require setting up connections each time  users want to publish or consume data. This means publishing the same  data multiple times to different end users. In sports, this often means  data ‘hops’ between upwards of four intermediaries before it reaches  exchanges, and multiple servers need to be used across geographies to  reduce latency and accommodate growing volumes.” In other words,  leveraging realtime sports data APIs for new commercial possibilities  requires significant manual integration to enable functioning with other  systems. Ably’s API Streamer comes to market as the first full  lifecycle realtime API management tool. In sports and elsewhere, its  ambition is to serve as a catalyst for industries being transformed by  realtime (sports data being one of the most prominent of these).

    When  it comes to how this works in practice, relevant to the sports  industry, information on a pass or a goal must be distributed through  multiple custom-built channels. With the API Streamer the infrastructure  requirements are the same (i.e. minimal) whether the data is published  out to 1 or to 100,000 end-user devices. This results in significant  cost-savings, as well as savings in engineering effort. Data owners have  more time and engineering capacity for turning data streams into  revenue streams, safe in the knowledge their realtime systems will  scale. The in-built interoperability of the Ably platform allows  developers to access data streams from a single place — data is more  usable for third-parties and has much further reach.

    Realtime tech and live official scores

    Tennis Australia (also powered by Ably) is one of the first sports bodies to set a data  performance target to consistently broadcast messages to 1.2 million  users within 500ms for the 95th percentile of users. According to Tennis  Australia, “The  Australian Open’s worldwide audience engage with the event via every  digital medium and format imaginable, with millions of users receiving  billions of realtime messages each second. Reliable realtime  infrastructure ensures Tennis Australia remains the primary source for  live scores, streaming videos, profile information, schedules and  commentary.” IMG, to use another example, has developed a real-time, shot-by-shot data collection system installed at all European Tour events, and powers a popular live scoreboard. Intel True View,  used by Premier Leagues, the NFL and the NBA, allows for virtual game  experiences similar to being on-pitch. However, the importance of  official data goes beyond this.

    While  the ‘Big Four’ sports data providers (Sportradar, Genius Sports,  Perform Group and Scientific Games’ Don Best Sports) frequently sign  multi-million dollar contracts with superleagues, debates are ongoing as  to whether the availability of scores in realtime could have had an  adverse affect on sport integrity. In a widely-publicised review the Tennis Integrity Unit linked the new availability of live scores at  lower tiers of the game with hikes in betting-related corruption. The  review’s authors put increased match-fixing instances down to live  official scores becoming available at lower-tier levels of the game,  where salaries are lower and participants more pliable.

    One  solution demands sports data providers work with regulators and  anti-corruption campaigners to create centralized systems where official  data is fast, reliable and robust enough to render illegal or offshore  data providers redundant. According to Sportradar, “due to our real-time  processing we have also all but closed the time lag between event  occurrence and event recording that allows courtsiders to flourish”.  Additionally, Sportradar’s Integrity Unit employs advanced algorithms to  detect suspicious activity in how data is used.

    Is sports ready for a realtime data rethink?

    According  to Matthew O’Riordon sports data transfer systems are in line for a  technical rethink if they are to support the combined demands of  connected fans, competitive UX landscapes and the need for watertight  official scores. Just as API management tools made it easy for companies  to exchange and share data outside their organizations, realtime API  Management tools make multiple streaming connections much easier to  manage and aggregate. According to him: “The future for sports data is  to avoid the intermediaries, avoid congestion, and look at licensing  models that allow data to be sent directly from the source to the  end-user (a business or consumer) without affecting existing commercial  agreements. What’s needed are closed systems guaranteeing capacity and  availability while provide the shortest possible transfer route.”

    To talk about the tech, use cases and scope for realtime API management tools — in sports and elsewhere — get in touch.