Your watch tracks heart rate variability, sleep quality, body battery, training load, training effect, recovery time, respiration rate, blood oxygen, and stress. Your bike computer records power, cadence, temperature, elevation, and normalised power. Your app calculates fitness, fatigue, form, acute training load, chronic training load, and a readiness score you never asked for.
You have never had more data. You have also never been more confused about what to do with it.
The Noise Problem
Data is only useful when you know what question it is answering. Most athletes collect metrics without a framework for interpreting them. The result is not insight. It is noise dressed up as information.
A readiness score drops from 78 to 61 overnight. What do you do? Skip the session? Reduce the intensity? Push through anyway? Without understanding what drove the change, and whether the change even matters for today's purpose, the number creates anxiety without direction. You end up making training decisions based on the metric of the day rather than the principles that actually govern adaptation.
This is the core problem. More data points do not produce better decisions if the athlete does not know which data points matter, which ones are redundant, and which ones are simply irrelevant to the question they should be asking.
What the Most Data-Rich Programme in the World Actually Prioritises
The Norwegian triathlon programme is arguably the most sensor-rich training environment in endurance sport. Portable metabolic analysers, muscle oxygen monitors, core temperature sensors, power meters across every discipline. They collect more physiological data per session than most athletes see in a year.
And yet, when asked what sits at the top of the decision hierarchy, Olav Bu's answer is unambiguous. Feeling. How the athlete perceives the session. Whether the effort matches the output. Whether today feels like what the numbers say it should feel like.
He has been explicit about this: despite the technology, despite the measurement precision, it still comes down to the feeling of the athlete. He never takes a measurement without first asking how the athlete feels. If the feeling and the measurement diverge, understanding why becomes the priority. The measurement does not override the athlete. It validates or challenges what the athlete already knows.
This is not anti-technology. It is technology used correctly. The sensors answer specific questions. The athlete provides the context that makes those answers meaningful.
The Diamond Shape of Development
Bu describes athletic development as a diamond shape. From novice to world class, you add tools, metrics, routines, and knowledge. The athlete grows by accumulating capability. But from world class to world champion, the process reverses. You remove noise. You strip away everything that does not directly serve performance.
Any new addition must replace something. It cannot simply be stacked on top. Decision fatigue is a real performance limiter, and every new metric, every new app notification, every new dashboard consumes cognitive resources that could be spent on training, recovering, or doing nothing at all.
Most age-group athletes operate on the wrong side of this diamond. They keep adding. A new recovery app. A new heart rate metric to track. A new readiness algorithm. Each addition feels like progress because it feels like control. But the accumulation of tools without the removal of noise produces a training environment that is more complex, not more effective.
What Actually Matters
For the age-group triathlete, the data that genuinely informs training decisions is a short list.
Heart rate at known intensities tells you whether your aerobic system is developing. If the same session produces the same output at a lower heart rate than it did six weeks ago, your engine is growing. That is a real signal. It does not require an algorithm to interpret.
The power-duration curve tells you the shape of your fitness across every duration that matters. It identifies both thresholds, reveals your metabolic profile, and tracks whether the whole curve is lifting or whether you are pivoting towards a single quality. Tested every six to eight weeks, it replaces the need for most other fitness metrics.
Perceived effort tells you whether today's training is hitting the intended target. If a session that should feel moderate feels hard, that is a signal to investigate. If it feels easy, your fitness is probably ahead of where you thought it was. RPE is free, requires no device, and when calibrated against objective data over time, becomes remarkably accurate.
Power and pace confirm that your intensity discipline is holding. They prevent the easy sessions from creeping up and the hard sessions from drifting off target.
That is the list. Heart rate. Power and pace. RPE. The curve. Everything else is either redundant, context-dependent, or answering a question the age-group athlete is not yet equipped to ask.
Less Data, Better Questions
The instinct to collect more data comes from a reasonable place. If data reveals something useful, more data should reveal more. But this logic breaks down in practice because the bottleneck is never the data itself. It is the framework for interpreting it.
An athlete with four good metrics and a clear understanding of what each one means will make better training decisions than an athlete with forty metrics and no interpretive framework. The first athlete knows what to look for, knows what normal looks like, and knows when something has changed in a way that matters. The second athlete is scrolling dashboards looking for something to feel concerned about.
Strip it back. Know what your key sessions are tracking. Know what normal looks like for you at those sessions. When something changes, notice it. When nothing changes, trust the process. The data should confirm what good coaching and honest self-assessment already suggest. It should not be the thing doing the thinking.
The data that matters is the data that changes what you do. If a metric never influences a decision, stop tracking it.