The Connected Athlete: The Role Data Science Plays In Athlete Management

The role of technology in elite sports performance continues to grow exponentially. Wearable tech, immersive reality, and video replay are just a few examples where technology is taking an increasingly dominant role in what is an incredibly exciting time for the entire elite sporting ecosystem.

Intel have recently invested in revolutionary data aggregation and amalgamation company Kinduct, a cloud based software that analyses data from wearables, electronic medical records, computer vision solutions and more to assess and make recommendations about physical human performance and injury risk.

Kinduct will be at the Football Innovation Summit in London next week (March 22/23rd) and leading a series of workshops with some of the leading clubs in European Football.

“The Connected Athlete: The Role Data Science Plays In Athlete Management”

We spoke with Kinduct CEO, Dr. Travis McDonough earlier this week to understand more...

S.I.N? - Tell us more about the technology you are using at Kinduct - How does this allow you to amalgamate and leverage data to create high performance insights and outcomes.

Dr. McDonough - Kinduct started as a company providing content like strength and conditioning tools and 3D medical animations to teams and we quickly learned that in order to grow in the industry we needed to begin ingesting athlete data, particularly as the wearable technology industry began to grow at a remarkable pace.  Kinduct grew quickly into a company that also provided a secure internet platform allowing clients to bring in and store in one location all the athlete data they collected. This was only part of the solution, we also needed to provide a set of reporting tools that would let our clients graphically display any of this information in ways that helped them understand the vast data sets they now had compiled.

The use of Application Program Interfaces (APIs) within the wearables and athlete management industry has allowed us to directly ingest athlete data as it becomes available which allowed our clients to access and view data when they need it allowing clients to leverage collected data sooner to make actionable decisions. Our reporting tools allow for the creating of unique reports specific for a team, player, game or a report can be created once and applied to each individual within the team.

These tools save hours of time for our clients but more importantly they provide visuals of data in ways that can alert team staff to changes in athlete data which if not acted upon may lead to an unnecessary injury. The ability to capture and be alerted to changes at the single metric level or applied to a calculated acute/chronic ratio are critical to making the most informed decisions for the good of the club and athlete.   

The next step for Kinduct has been to partner with companies like Intel where we can combine Kinduct’s platform with Intel’s Saffron Natural Intelligence Platform™ and Trusted Analytics Platform (TAP) technology to take all the data a client stores within Kinduct and make recommendations such as predicting the potential risk for player injury with higher accuracy by combining continuous learning and persistent memory with the power of computing. This has never been done properly before and we are excited to be working with the best technology companies in the world to bring the best athlete management platform to the European Football Market.

Kinduct exists as a central repository for athlete information, allowing a sports organization to efficiently view data collected in various ways in a single location and spend more time taking action on insights and less time searching between various silos of information. By giving time back to team staff so they do not have to log into different individual platforms for in-game tracking, practice GPS, heart rate, weight room performance, etc. etc, they can spend more time pulling out nuggets of actionable truths and delivering programming. Kinduct achieves this by collecting data through a variety of means including API integrations, custom data imports and direct data entry. Once the data is collected, Kinduct's suite of reporting and visualization tools, along with our scoring, triggering and alerting features help identify key pieces of information. Finally, Kinduct provides the means to take action through strength and rehab programming builders supported by libraries of video content. The next step for Kinduct is to continue to grow our data science capabilities to help teams identify insights in their data quicker and more effectively with the end goal of shortening the amount of time a team needs to spend sifting through tables of information to allow them to focus on making their athletes better.

S.I.N - You have worked extensively across North American sports - what are the major challenges you've overcome, how has the technology driven direct performance outcomes on the field/court/pitch and how do you see these translating into the European Football Market?

Dr. McDonough - The foundation of Kinduct is to make athletes better by deriving from the data collected at the individual level actionable insights that are specific to a league, sport and team. Kinduct as a company has been able to grow with our clients and as we expanded to work with each new market we feel we are well prepared to adapt to any specific needs that the European football market brings.

The explosion in wearable companies providing unique products that in some cases are sport specific has led us to adopt an aggressive approach to be able quickly set up integrations with 3rd party devices so that we streamline the process of bringing in the data our clients want. This is important as there will be many European wearable companies that we will need to integrate with that are not available or used in North America. This puts Kinduct in a strong position to set up integrations quicker leveraging our existing integration architecture.  We take pride in the security protocols we have implemented in North America and know that we are able to adapting to different privacy laws that may exist within European countries.  The inputs that come into Kinduct are staggering and we have had to grow and respond quickly to this fast paced wearable industry. The next step is for us to provide a tool set that proactively reacts to data brought into a client's platform providing them with actionable tasks rather than simply remain as a storage and visualization platform.

A unique challenge we have had to overcome early on was to be able to provide the most effective solution for a single professional sports organization, like an NFL or MLB team which is in many ways a single silo of athletes, while also providing a platform that can be used by a National Olympic Organization which have many different silos of athletes, coaches and staff all while maintaining a structural organization that lets the client determine which sports staff within an organization need access which athlete data.
We know that football clubs are very large complex organizations and we know that our platform will be able to handle the various complex hierarchies that exist within the European football market.

The Kinduct platform works best in environments where multiple stakeholders all get involved in the process, helping streamline communication and collaboration between staff.

What sets Kinduct apart and will allow us to seamlessly work with the European football market is our industry leading level of client support. Each team or organization is assigned a highly experienced Kinduct Client Success Manager who works directly with client staff to build their Kinduct platform from the ground up taking each organization's specific needs so that we can provide a solution for each team to make their athletes better knowing how that organization works.

S.I.N - What are your predictions for the future role of predictive analytics in elite sports performance? Do you see these replacing or merging with existing methodology and technology?

Dr. McDonough - Predictive analytics is an area that will continue to grow and evolve as an effective part of the sport performance staff's tool box. However, it is important to keep in mind that the value is not with the technology, but with those who use the technology.
Predictions, like the other tools in the tool box, will only be a single voice and not always a fit for every scenario. The filtered down data that is provided by various wearables still needs to be validated by the research community, until this happens it will be hard for predictive analytics to fully replace what is currently used. As the reliability and validity of data collected, while also validating the predictive models and how they are implemented, continues to improve and be better understood there will be a merging with existing methodology and technology. There remain opportunities for growth in the research and application of appropriate actions and plans to capitalize on or mitigate the effects of the prediction. This is where effective predictive tools become part of a cohesive strategy, helping guide organizations to healthier athletes and improved performance.

Predictive analytics will also need to learn how to ingest new sources or forms of data that clients will want to take advantage of and so there must be a learning phase with each new data source that comes to the market.

Another challenge is that data is now very athlete specific and any predictive analytics applied to this data must be able to work at the individual level, which will be a challenge to overcome.

S.I.N - "The Internet of Things" is a buzzword across several industries at the moment? What exactly does this mean and what are the biggest restrictions before it becomes fully realised in sports performance

Dr. McDonough - The Internet of Things (IoT) has a few interpretations, but if I were to summarize the IoT in the athlete management industry it would be the inter-connecting of physical devices or mobile apps to provide a 360 degree view of an individual athlete's data that can be used to identify specific factors that will affect player performance. In order to be competitive organizations must continue to be involved in using the latest technology, research and coaching tools so they can determine what is an optimal skillsets, base physiological and mental level required to mitigating the effects of training. As more technology and information becomes available to evaluate an athlete, such as brain training and incorporating insights from genetic testing and hormone tracking, the full picture of athlete performance is within reach.

I would not necessarily say there are restrictions, just leagues, organizations and player associations being cautious about understanding the ramifications of how all this data can be used in ways that may negatively impact athletes, such as during contract renewal, rather than simply thinking of athlete data as a means to improve athletic performance. This leads into the discussion around of ownership of player data.  Should an athlete's data follow that player when the move from one organization to a new one or when training or playing for their country? The ultimate goal is to make an athlete better, and as inter-connecting devices gather more and more data on athletes there may need to be policy decisions that the FIFA and regional governing bodies need to agree upon so that data can be used to make athletes better, or mitigate injury, while also treating the athlete as a human within a organisation.