Power Profile explained: how Moveee places you from sprinter to diesel

The Power Profile page takes every ride with power that you have synced and turns it into a handful of numbers and one shape. This post explains where each number comes from, in the order the page shows them, and what "sprinter" or "diesel" actually means when Moveee draws it. Nothing on the page is measured in a lab. Everything is estimated from your best efforts, and the page tells you how much to trust each estimate.

What it is built from

For every ride with power, Moveee computes your best average power over a set of durations, from 5 seconds up to an hour and beyond. The page then aggregates the best value at each duration across the rides in the window you have chosen: the last 90 days by default, a rolling window, one season, or all time. The result is a power–duration curve, made monotonic so that a longer best implies at least that power for shorter windows. Each point on the curve links back to the ride that produced it. If you have set your weight, the curve is drawn in W/kg with rider-category reference lines behind it, from World Tour down to Cat 4–5; without weight it is plain watts.

Critical Power and estimated FTP

The card titled "Critical Power & estimated FTP" says it is "fitted from your best 3–30 min efforts", and that is exactly what happens. The Critical Power model says power at exhaustion follows P = CP + W′/t. Regress your best powers against 1/t and the intercept is CP, the slope is W′, the work you can do above CP before you stop. Moveee only fits efforts between 3 and 30 minutes. Shorter efforts are excluded because they are dominated by anaerobic capacity and systematically pull CP upward, a protocol effect that has been documented carefully. Longer efforts are excluded because almost nobody in a normal ride history has ridden 40–60 minutes truly flat out, and a sub-maximal long effort drags CP down.

Estimated FTP is then CP × 0.97. Critical Power sits a little above the roughly hour-long FTP in trained riders; a study comparing the two found CP about 7 W higher on average, and the 3% correction is a modest nod to that rather than a claim of precision. We wrote about the difference at length in Critical Power vs FTP. The W′ card shows the slope in kilojoules.

The fit comes with a confidence label. High needs at least four efforts, including one of 20 minutes or more, sitting on a tight line (R² ≥ 0.97). Medium needs three efforts and R² ≥ 0.90. Anything else is Low, and the tooltip names what fell short: "fewer than 4 efforts", "no 20 min+ effort", "the line isn't tight enough". The honest caveat is in the code itself: the fit cannot tell whether the efforts were maximal. A Low label most often means you simply have not gone all out recently.

Thresholds and FatMax

LT2, the second lactate threshold, is set equal to FTP; it is what FTP was designed to approximate, and the two agree more closely in fitter riders, so the page attaches a confidence that rises with your W/kg. LT1, the aerobic threshold, is estimated as 66–75% of FTP, scaled by fitness: the higher your FTP per kilogram, the closer LT1 sits to FTP, and 70% is used when your weight is unknown. Both are labelled estimates because only a blood-lactate test measures them, and the validation studies show FTP tracks lactate markers well in trained cyclists but drifts in the less trained.

FatMax is anchored to that LT1 estimate rather than to a blind percentage of FTP: the band runs from 80% of LT1 up to LT1, with the fat-oxidation peak placed at 90%. That follows the finding that maximal fat oxidation occurs at a moderate intensity just below the aerobic threshold, and that it is a broad plateau rather than a point. If you have a maximum heart rate on file the card also shows the classic 68–80% of max HR band.

VO₂max

The page estimates VO₂max from your best 5-minute power and your weight, using regressions derived from 580 competitive cyclists tested in a lab. There are separate equations for men and women, and a pooled one for riders who have not said; the pooled sample was 86% male, so a woman scored on it reads several points high, which is why the profile asks. The same study found the older ACSM leg-cycling equation underestimated VO₂max in this population by 11–15%. Your 5-minute best stands in for the lab's maximal aerobic power, so the estimate is only as good as your hardest recent 5 minutes.

The shape: Power Index and rider strengths

The "Power profile" card is a spider chart with seven spokes: 5 s, 15 s, 30 s, 1 min, 5 min, 20 min and 60 min. Each spoke shows a Power Index from 0 to 100, which is your W/kg at that duration placed between a floor (recreational) and a ceiling (a strong Cat 1 amateur). At 5 seconds the floor is 9 W/kg and the ceiling 20; at 60 minutes they are 1.8 and 4.8. The scale is deliberately amateur-focused, so a good club rider sees numbers in the middle rather than a dot at the origin.

The card's own subtitle says it: "the shape tells you whether you're a sprinter, an all-rounder or a diesel." A spiky front, high on 5 and 15 seconds and low at 20 and 60 minutes, is the sprinter's shape. A full back, low at the front and high at 20 and 60 minutes, is the diesel's. A bulge at 1 and 5 minutes with an ordinary back is the puncheur. A round shape is the all-rounder. This is the same idea as the record power profile used to characterise elite riders, scaled down to the people who actually use the page.

Moveee does not stamp a label on you. There is no "you are a puncheur" line, and we chose not to add one, because the cut-offs would be arbitrary. Instead the "Rider strengths" card scores four energy systems, Sprint (5 s), VO₂ max (5 min), Threshold (20 min or your FTP) and Endurance (60 min, falling back to 20), on the same floor-to-ceiling scale, and the "Your strengths and weaknesses" card names your highest and lowest with a sentence each. Threshold as the weakness reads "you fade on sustained 10–30 min efforts" and the fix suggests "threshold blocks (2×20, sub-threshold 5×6) and sweet-spot work". That weakest system is also what the plan wizard targets when you leave the intensity emphasis on Balanced, so the shape is not decoration; it decides what kind of hard days you get. What your power curve says about you goes further into reading the shapes.

Where you rank

Two estimates from the same FTP: a weight-aware one in W/kg and a pure-watts one, each placed on a Coggan-style ladder from World Tour to Cat 4–5. The men's ladder is hand-set; the women's is derived from it by the ratio between the women's and men's columns of the classic classification table, about 0.83 at 5 seconds and 0.87–0.88 at FTP, so the two never disagree about the size of the gap. Without a weight you only get the watts column, and the page says "Add weight for this lens".

Glitches, and why your best 20 minutes vanished

Power meters drop out and spike. A stuck-high reading on a trainer can produce a 20-minute "best" you never rode, and one bad file will quietly set your estimated FTP. So three nets run by default under "Hide likely glitches".

• Per ride: no effort may beat the world-record W/kg for its duration (25 W/kg at 5 s, 6.8 at an hour), and a short effort that towers over the next-longer one faster than power can physically fade is a spike.
• Per rider: a genuine personal best is crowded, with other rides just behind it; a glitch stands alone. For each duration with at least five rides, an effort with no supporting cluster below it is dropped.
• Per duration: once a duration has eight or more rides, population outliers go.

The Sensitivity menu (Cautious, Balanced, Aggressive) sets how close "near" is and how many companions a peak needs; Aggressive will also peel two bad readings that landed near each other, at some risk to a real one-off standout. Dropped efforts are listed under the curve with a note that they are "far above your usual cluster", and unticking "Hide likely glitches" puts them all back. You can also exclude a ride by hand under "Manage rides in this power profile". Changing sensitivity or excluding rides updates the page at once; the stored profile that the plan reads is refreshed with the Rebuild action ("Refreshing your power curve — re-checking your rides"), which is rate-limited.

Adding a lab test

The "More" menu has "Add lab test (PDF)". Upload a metabolic or lactate test report and it is read in the background: test date, lab, VO₂max, LT1, LT2, FatMax with fat-oxidation rate, gross efficiency, and the lab's own zone table. You review the extracted values before they are saved. A confirmed test then appears in the time filter as a snapshot; select it and the page switches to a "measured in the lab" view instead of the ride-estimated curve. The extractor normalises terms and units from reports in any language (many of the labs our riders use write in Slovenian), but layouts vary, and the review step exists because extraction is not perfect.

Honest limits

• You need a power meter or a smart trainer. Heart-rate-only rides contribute nothing here. Without rides the page says "Connect Strava to see your real power curve".
• Estimates track effort, not fitness. A quiet month of endurance drops your 90-day CP fit even if you are fitter. Switch the window to a season before concluding anything.
• Weight matters more than it seems. W/kg, the category ladder, VO₂max and every Power Index need it; an old weight skews all of them.
• The plan uses your profile FTP. Sessions are written as %FTP and converted with the number on your profile. The page shows the gap between that and the estimate; it does not overwrite it. Decide for yourself, and read why your FTP went down before you do.

Open your Power Profile, hover the Critical Power confidence badge, and if it says Low, go and ride a hard 20 minutes. That one effort will do more for every number on the page than any setting.