Athletes have never known more about themselves.
A smartwatch can monitor heart rate and recovery. Another app tracks sleep. Practices are recorded on phones, wearables measure workload, and increasingly sophisticated tools can break down movement and technique in ways that once required specialized equipment.
The problem is that much of this information still lives in different places.
Victoria Drazdova, a former professional athlete, researcher and head coach at VSA, sees a much bigger opportunity than simply adding another generation of wearables or performance apps.
“Right now, an athlete can know how they slept, what their heart rate was, how much they trained and what happened technically in practice, but all of that information is usually sitting in different places,” Drazdova said. “You might have sleep in one app, recovery somewhere else, hundreds of training videos on your phone and feedback from your coach. The real breakthrough will be bringing those pieces together so you can understand the athlete as a whole.”
Connecting the whole athlete
That kind of integration could fundamentally change what personalization means in sports.
A system that tracks only workload can say something about workload. Add recovery, sleep, technical development and an athlete’s longer-term performance history, and the picture becomes considerably more useful.
Instead of opening several dashboards, coaches and athletes could begin seeing relationships between information that is rarely examined together today.
A difficult training session, for example, may have little to do with technique. It could follow several nights of poor sleep, an unusually heavy training block or a demanding competition schedule. The reverse is also possible: An athlete may feel physically strong while a technical pattern has been deteriorating gradually for several weeks.
Those relationships are difficult to spot when every part of an athlete’s history exists in isolation.
Some of that thinking is already visible at the highest levels of sports. The NFL and Amazon Web Services developed the Digital Athlete, which uses artificial intelligence and machine learning to build individualized views of players’ experiences. Other elite sports programs are experimenting with computer vision, sensor technology and automated video analysis.
The individual tools matter less than the direction they point toward. For years, sports technology was largely about producing better measurements. The next competitive advantage may come from making those measurements useful together.
Less data, better decisions
Artificial intelligence could eventually do much of that connecting in the background.
Its most valuable role may have little resemblance to the heavily marketed idea of an “AI coach.” Instead, it could sort through large amounts of information, recognize relationships and bring a small number of observations to the surface for an athlete or coach to consider.
“The future isn’t a dashboard with 200 numbers,” Drazdova said. “Nobody needs that. If AI is going to be useful in sports, it has to take all of that complexity and tell us what is actually worth paying attention to.”
That could also change when coaches intervene.
Most training decisions today are reactive. An athlete reaches a plateau, performance becomes inconsistent or a technical problem becomes obvious, and the program is adjusted.
With enough historical context, future systems could potentially recognize smaller changes before they become larger ones.
The prediction does not require software to know exactly what will happen next. It may be enough to notice that an athlete is no longer responding to training in the same way, that a familiar technical issue is appearing more often or that performance repeatedly changes under a particular combination of workload and recovery.
“That’s where I think it becomes genuinely interesting,” Drazdova said. “Not when AI gives us another score, but when it notices a relationship we might otherwise miss and gives us a reason to look closer.”
Bringing sophisticated analysis to more athletes
There is an economic consequence as well.
Professional organizations can already surround athletes with people who manually connect many of these dots. Coaches speak with performance staff, conditioning specialists review workload, medical teams monitor recovery and analysts study performance.
Most developing athletes will never have that infrastructure around them.
Software cannot recreate an Olympic or professional support team, but it could eventually perform some of the coordination that currently requires one. As computer vision improves, sensors become less expensive and processing becomes more powerful, selected forms of sophisticated analysis are likely to become available at a fraction of today’s cost.
That could reverse a familiar pattern in athlete development.
Traditionally, the more successful an athlete becomes, the more sophisticated the support around that athlete becomes. Better analysis, specialist expertise and multidisciplinary performance teams often arrive only after an athlete has demonstrated elite potential.
“We’re used to the best technology reaching athletes after they become elite,” Drazdova said. “I think we’re going to see some of that happen much earlier. Young athletes will grow up with access to information that previous generations only received much later in their careers.”
That does not mean technology will manufacture elite athletes or remove the years of work required to become one. Physical development, repetition, coaching and competitive experience cannot be compressed into an algorithm.
What better information may reduce is some of the trial and error that takes place during those years.
Why coaches still matter
Technology could also change the economics of coaching itself.
Sports instruction is still largely sold in units of time: an hour with a coach, a session with a specialist or another appointment added to an athlete’s schedule. Expertise is inherently difficult to scale under that model because even the best coach has a finite number of hours available.
Part of that relationship may eventually become asynchronous. Technology could handle some of the observation, comparison and organization that takes place around training, allowing athletes to use specialist time for problems that genuinely require specialist judgment.
The individual lesson would not disappear, but the reason for booking one could become more specific.
For coaches, that shift could make experience more valuable rather than less.
Recognizing that something changed is only the beginning of a coaching decision. Understanding why it changed, whether it matters, what should be addressed first and how a particular athlete is likely to respond are considerably more difficult questions.
A system might recognize a relationship between poor sleep and a change in performance. It cannot fully understand whether an athlete is anxious about an upcoming competition, losing confidence after an injury or simply having a difficult week.
More information can improve a decision, but it cannot supply all the human context behind one.
This is one reason the most successful sports technologies may be those that eventually become the least visible. Athletes may stop thinking about whether they are “using AI” in much the same way people rarely think about the algorithms operating behind navigation or search. The intelligence may simply become part of the environment surrounding training.
Drazdova said some of these ideas are already informing the technology-assisted training being explored through her work at VSA. Her broader prediction, however, extends beyond any single platform or sport.
Who owns an athlete’s data?
If different parts of an athlete’s development begin to exist within the same system, the next question will be who controls that information.
A young athlete could eventually accumulate years of technical development, performance history, workload and recovery data. Changing coaches or organizations would no longer necessarily mean leaving that context behind.
A new specialist could begin with a clearer understanding of what had already been tried, where progress accelerated and which problems repeatedly returned.
That would make athlete data increasingly valuable and potentially contentious.
Who owns a long-term performance profile? Can athletes take it with them? How much should be collected from children, and who should be allowed to see it?
Sports organizations have spent years learning how to generate more data. The next decade may force them to decide who that data belongs to.
“If you’ve spent years building your athletic history, I think you will eventually expect that information to move with you,” Drazdova said. “Changing a coach shouldn’t mean that years of useful context simply disappear.”
There are also reasons to be cautious.
More measurement does not automatically create better athletes, and constant analysis could become counterproductive, particularly in youth sports. Sleep scores can be misleading, wearable data can be imperfect, and a young athlete does not need every practice transformed into a performance report.
The strongest systems may therefore be defined as much by what they choose not to show as by what they can measure.
The goal is not to produce hundreds of metrics. It is to identify the handful of things that matter at a particular moment and leave the rest in the background.
Athletes will still have to do the physical work. They will still need coaches, facilities, repetition and the patience to improve slowly. No software is likely to change that fundamental reality of sports.
What could look surprisingly dated a decade from now is everything surrounding that work: checking one app for sleep, another for recovery, searching through a phone for old training footage and trying to remember what a different coach observed months earlier.
The next generation of sports training may not be defined by having more technology. It may be defined by finally making all of that technology understand the same athlete.