Once every stroke of a piece is plotted, the chart stops being a summary and starts being a transcript. That takes some getting used to — most erg data is so heavily averaged that people are not practised at reading the raw version.
There are four patterns worth learning to spot. Each one has a distinct cause and a distinct fix.
1. The settle
What it looks like: the first eight to fifteen strokes sit well above everything after them, then the trace drops onto a level that holds.
Almost everyone does this, and a small settle is fine — the flywheel is coming up to speed and the first strokes genuinely require more work. A large one is not the flywheel. It is going out too hard because it feels easy, and it is paid for somewhere after 1200m.
Fix: decide the target split before the piece starts and hold it for the first 500 even when it feels restrained. If the first ten strokes are 40 W above the piece average, that is a pacing problem, not a fitness one.
2. The drift
What it looks like: no single failure point. The whole trace tilts gently downward from start to finish.
This is honest fatigue, and it is the pattern you want if you are going to have one. The size of the tilt is your fade, and it is trainable — it responds to volume and to threshold work far more than to strength work.
Fix: nothing acute. Track the slope test over test. A drift that gets shallower across a block is the block working, even if the finishing times have not moved yet.
3. The collapse
What it looks like: the trace holds, then a cluster of strokes falls off a cliff — often three to six of them, usually somewhere between 1000m and 1500m — and then recovers, partially.
This is the pattern that a split table hides best. Averaged into a 500m split it appears as a second or two, indistinguishable from ordinary fatigue. Stroke by stroke it is obvious, and it is specific: something failed briefly and got picked back up.
Usually it is concentration, or the point where the drive shortens and the rate compensates before the athlete notices.
Fix: the diagnostic is stroke rate. If rate climbed while power fell across those strokes, the drive shortened — a rate-capped repeat of the same piece, holding length at a fixed rate, addresses it directly. If rate stayed flat and power fell, it is more likely pacing or attention.
4. The scatter
What it looks like: no trend at all, but every stroke sits 30–60 W away from its neighbours. The trace looks like noise.
Two athletes can average identical watts with completely different amounts of scatter, and only one of them is repeatable. Consistency is the quality that decides whether a good test can be produced again on a bad day.
Fix: it is a technique and rhythm problem far more often than a fitness one. Steady-state work at a fixed rate with a target power band, rather than intervals, is what moves it.
Why the target line matters more than the average
A power-per-stroke chart with a target line drawn across it answers a different question from a chart of averages. The average tells you what happened. The target line tells you how much of the piece was actually spent doing the thing that was prescribed.
An athlete who averages exactly the target power by spending half the piece well above it and half well below it did not do the session. That distinction is invisible in every summary metric, and it is completely obvious the moment the strokes are plotted individually.
This is the level of detail skiergonomy records — every stroke of every session, captured from the wrist. If you want the arithmetic behind the numbers first, the pace calculator and the pace chart are free and need no account.