Turn data into performance: understand internal and external load, track threshold, efficiency and durability, and steer intensity distribution data-based.
Data without interpretation are just numbers — at elite level it is decided by who asks the right questions of their data. The direct answer: measure the quantities that truly determine your endurance performance — VO2max, lactate threshold, efficiency and, over ultra distance, durability — and steer your intensity distribution and load from them. Everything else is a data graveyard. Many ambitious riders' mistake is collecting terabytes of rides without ever translating a metric into a decision.
The second key is separating internal and external load. External load is what you produce (watts, kilometers, elevation); internal is what it costs your body (heart rate, session-RPE, strain). The consensus in training science is to monitor both, because only their ratio makes adaptation and overload visible. The same wattage can feel effortless one day and grueling the next — and it is exactly this difference, not the wattage itself, that tells you how your form stands. Whoever looks only at the external side reads their own training half-blind. This guide shows you how to translate your performance data into decisions — data-driven, but never data-blind. (The control models are transferred from endurance training science to gravel.)
Focus on the determinants that count: threshold power, a VO2max proxy, efficiency and durability. Define a measurement method and a test interval for each. Do not collect everything, but what you can translate into training decisions — every metric needs an "if X, then Y" consequence.
Capture external load (watts, kilojoules, elevation) and internal load (heart rate, session-RPE) separately. Their ratio is the actual signal: if internal load rises at the same external load, it points to fatigue or insufficient recovery. If internal drops at the same external, your form grows.
Check data-based whether your real distribution matches your planned polarized one: around 80 % below the first, 15–20 % above the second threshold, little in between. A time-lapse over weeks reveals the most common elite mistake: the creeping drift into "grey" medium pace, which creates much fatigue with little adaptation.
Establish reproducible reference tests: same climb, same power, measured heart rate and exertion — falling internal cost at the same power means rising efficiency. For durability compare your performance in the last third of long sessions with the first: if it stays more stable, your fatigue resistance has improved. Additionally you can check pedal effectiveness via pedal data without falling for the "round stroke" myth.
Monitor training monotony and strain via session-RPE; high load plus high monotony is an overload signal. Interpret all data in the context of your honest performance classification — the same number means different things for a tier-3 and a tier-5 athlete. Only context turns numbers into decisions.
Use data for early overload detection, not as a driver to overdo it: a rising strain at rising monotony is a braking signal, not an incentive. A suddenly elevated resting heart rate, falling heart-rate variability or dropping performance at the same effort can indicate overtraining or infection — then reduce load, if necessary get medical clearance. Perform maximal tests only well-rested. Data do not replace body feel, they complement it.
Build a weekly "one-glance dashboard" with exactly five quantities: planned versus real intensity distribution, chronic load, acute load, monotony/strain and an efficiency or durability marker from your reference ride. No decision needs more. Whoever instead watches dozens of metrics ends up seeing none — the art of elite analysis is leaving out, not collecting. Every number on the dashboard must be able to trigger an action.
FTP is your threshold power and the anchor for training zones and load analysis. Via it you define your intensity distribution and normalize rides for comparison. Complement it with internal load measures like heart rate and session-RPE, because only the ratio of external to internal load shows whether you are adapting or fatiguing.
Evaluate your time in the training zones over weeks: with a polarized distribution around 80 % lies below the first and 15–20 % above the second threshold, little in between. A time-lapse reveals the creeping drift into "grey" medium pace — the most common mistake of ambitious riders, which creates much fatigue with little adaptation.
Compare your performance in the last third of long sessions with the first third, at the same effort. If performance stays more stable in the tired state or internal exertion for the same power rises less, your fatigue resistance has improved. Standardize the conditions so the comparison stays meaningful over the weeks.
Pedal data show you pedal effectiveness, but beware the myth: active pulling raises mechanical effectiveness but lowers gross efficiency. Use the data to spot gross asymmetries and inefficient patterns, not to optimize for maximum "roundness". The pushing, freely chosen stroke remains energetically the most efficient.
Gravel-Biking: Sports Psychology