Why late-session efforts need context
What a power curve leaves out
A season-best five-minute effort could happen early in a short session, while a race might demand a hard effort after hours of running or riding. A chart that keeps only the best result at each duration discards this context. A runner can compare late-run pace at a similar effort, taking the terrain and conditions into account; a cyclist can compare power.
The cycling experiment draws two example power curves. One represents fresh efforts. The other applies a selected reduction to show how a late-ride comparison might look. The reduction is user-chosen, not a physiological prediction. Compare the same effort duration in both curves to explore why fresh bests alone may not describe the demands of a long event.
Solid: fresh example. Dashed: chosen reduction. These are hypothetical curves, not a fitted fatigue model or a predicted training response.
What a power curve leaves out
A season-best five-minute effort could happen early in a short session, while a race might demand a hard effort after hours of running or riding. A chart that keeps only the best result at each duration discards this context. A runner can compare late-run pace at a similar effort, taking the terrain and conditions into account; a cyclist can compare power.
The cycling experiment draws two example power curves. One represents fresh efforts. The other applies a selected reduction to show how a late-ride comparison might look. The reduction is user-chosen, not a physiological prediction. Compare the same effort duration in both curves to explore why fresh bests alone may not describe the demands of a long event.
Comparing efforts after a long ride
Two hours at different powers can accumulate very different mechanical work. A field study of professional cyclists examined maximal mean power at different levels of accumulated work, and investigated how repeatable those observations were.
That is a reason to preserve work-before-effort alongside duration and watts. It does not give every rider one universal fatigue threshold.
What affects a late-ride effort
Select a duration and change the late-ride reduction. The visible gap is an arithmetic consequence of the setting. It is not an estimate of what a particular training plan would improve.
In real files, compare efforts that were actually attempted hard, with attention to device, terrain, pacing and conditions. A lower late-ride number could reflect the absence of a maximal attempt.
Looking at your own ride data
Research in amateur road cycling also examines power declines and durability. The topic is broader than one best-power number and remains dependent on how efforts and fatigue are measured.
For your next analysis, locate the effort and inspect what came before it. Cyclists may have accumulated mechanical work; runners can consider elapsed time, distance, elevation and the intensity of earlier running. These are different measures, not interchangeable fatigue scores. Keep the original trace close.
Why the effort before the effort matters
A fresh best effort shows what was recorded under one set of conditions. A long event adds hours of prior work, heat exposure, changes in fuel availability and accumulated strain before the decisive effort arrives. Those factors can interact, so late-session performance cannot be explained by one number alone.
For example, compare a five-minute climb early in a ride with another after several hours. Even if both are five minutes long, they are not identical tests unless the rider actually attempted them similarly and the terrain and conditions were comparable.
Why accumulated work is useful but incomplete
On a bike, mechanical work in kilojoules provides a way to describe how much work was recorded before an effort. Two hours at 100 W is 720 kJ; two hours at 200 W is 1,440 kJ. Equal duration does not necessarily mean equal prior work.
Intensity and the pattern of work still matter. A review of cycling studies found greater reductions in subsequent power after some higher-intensity protocols despite less accumulated work. That makes kJ useful context, but not a universal fatigue meter or an interchangeable measure for running.
A useful comparison for runners and cyclists
Choose the same effort duration or route segment, and compare attempts that were actually hard enough to be informative. Record what happened beforehand: elapsed time, intensity and conditions for either sport; on a bike, recorded mechanical work can add another piece of context.
For runners, pace is affected by slope, wind, surface and GPS quality. For cyclists, a late-session power result is affected by the terrain and whether a hard effort was attempted. A missing late best can mean there was no opportunity or reason to produce one.
What training can—and cannot—be inferred
Durability describes how characteristics change as prolonged exercise continues, but studies use different tests and definitions. A smaller decline in one protocol is not automatically the same as being better at every long event.
Use repeated, comparable observations to identify a question for training: does the goal event demand a sustained effort after a long lead-in, repeated accelerations, or holding pace late in a run? Choose observations that resemble that demand rather than treating one fatigue threshold as universal.