Which ROUVY FTP test should you take, and what the number rewrites
ROUVY publishes four FTP tests in one tab: a Ramp Test, a Ramp Test Lite for riders under 70 kg, and two 20-minute tests that differ only in warm-up length [ROUVY FTP Tests]. Which you pick matters less than riders expect. What matters is what ROUVY does with the result. Your FTP is not a score you file away here. It is the setting that recalculates every workout in the library [ROUVY Workout Library], and it sits twice inside every Training Stress Score.
By Jim Camut · Former pro & ex-Bruyneel Academy racer
Updated Sep 17, 20264 chapters7 citations
The four tests in ROUVY's FTP tab
Two are ramps and two are 20-minute efforts [ROUVY FTP Tests]. The ramps compute a number for you. The 20-minute tests do not: you type a target FTP in first, ROUVY adds 5% to it, and that becomes your target power. One shape measures, the other confirms a guess.
The Ramp Test adds 20 watts every minute and stops on its own when you can no longer hold the target [ROUVY FTP Tests]. It runs about 40 minutes including a warm-up and a cool-down, both skippable while the testing phase is not. Ramp Test Lite is the same protocol for riders below 70 kg. At the end ROUVY shows the new FTP and offers to write it into your profile. One decision, one button, which is why the ramp suits a rider with no number at all.
The 20-minute tests work differently and the difference is easy to miss. You enter a target FTP, ROUVY adds 5% to your entry, and that sum is the power you hold for the 20 minutes [ROUVY FTP Tests]. ROUVY also defines FTP as 95% of the average power over a 20-minute effort. Ride 5% above your entry, take 95% of it, and the test hands back close to what you typed. ROUVY says as much itself: start at 200 watts if you are unsure, and restart with a different value if that proves wrong.
So the decision is narrower than a four-item menu suggests. With no trustworthy number, ride a ramp, the only shape that produces a figure from nothing. With a number you trust, take a 20-minute test and enter your current FTP rather than an optimistic or a cautious one. Which protocol tells more truth is not a ROUVY question: a ramp tends to read high for anaerobically strong riders, and the trade against a shorter effort has its own guide.
On ROUVY the number is a setting, not a result
ROUVY states it plainly: workout intensities are automatically calculated as percentages of your FTP, and when your FTP changes your entire workout library adjusts so targets always match your current ability [ROUVY Workout Library]. The figure you accept on the summary screen rewrites every session you have not ridden yet.
It reaches the load ledger as well as the watts on screen, and ROUVY publishes the formula that shows how. ROUVY publishes it as duration in seconds multiplied by Normalized Power and by Intensity Factor, divided by FTP times 3,600, and the whole thing multiplied by 100, with Intensity Factor itself being Normalized Power divided by FTP [ROUVY TSS and IF]. An hour at threshold scores 100, which is what the hundred is for. FTP appears twice, so the score moves with the square of the error: an FTP set 10% too low makes an identical ride score roughly 23% more TSS.
There is a mid-session escape and its shape is worth knowing. The resistance buttons raise or lower your FTP for that workout only, and maximum power is limited to 1.5 times base power [ROUVY Workouts]. One session can be walked down, or lifted within that ceiling, but neither press changes the setting that produced the error. How often to repeat a test is answered by our guide on how often to test FTP. The ROUVY part is that a test here is a write, not a free reading.
The power source is part of the test
ROUVY states that it primarily searches for the power meter and does not care about the trainer, and that the accurate data comes from the power meter no matter how resistance is controlled [ROUVY Power Source]. Whichever device ROUVY chose on test day is the device your FTP belongs to.
A power meter and a smart trainer are independent instruments and they will not agree. Bouillod and colleagues compared three popular meters against an SRM crank and measured the Stages unit 5.1% low in the heavy-intensity zone and the Garmin Vector 4.5% low in the same zone [Bouillod et al. 2017]. Gaps that size are larger than a season of adaptation. So test with the power meter paired, then ride with it paired. Leave it off, or pair a classic trainer with a speed sensor so ROUVY calculates virtual power from a speed curve [ROUVY Power Source], and your results stop being one series.
ERG is the other variable, and you can switch it on or off at any point inside a structured workout [ROUVY Workouts]. In a ramp that matters less than it sounds, because the ramp is self-terminating: the target climbs every minute until you cannot hold it and the test ends itself [ROUVY FTP Tests]. In the 20-minute tests you can increase, reset or decrease resistance as you go, so the pacing is yours. What ERG does on a filmed route is covered in our guide on ROUVY ERG mode on routes.
When the number comes back lower
Do not accept a lower FTP on the day it appears, and do not wave it away either. A drop has several plausible causes and only one of them is lost fitness. Work through the instrument, the protocol, the size of the drop, and the state you rode it in, in that order, before you touch the setting.
Instrument first, because it is cheapest to check and largest in effect. Was the same power source paired for both tests [ROUVY Power Source]? A swap between a crank meter and a trainer, a firmware update, or a skipped calibration can account for the whole drop [Bouillod et al. 2017]. Protocol second: a ramp and a 20-minute effort are different measurements and the number moves between them at identical fitness. If this was a 20-minute test, look at what you typed in, because a cautious entry lowers the power you were asked to hold.
Then size the drop against the noise. Hopkins and colleagues meta-analysed 101 studies and put the typical error of constant-power cycling tests lasting one minute to three hours at 0.9 to 2.0%, and peak power in an incremental test at about 0.9% [Hopkins et al. 2001]. On a 250-watt threshold that is 2 to 5 watts. Error ran about 1.3 times larger in non-athletes, and larger again on a first attempt: performance rose 1.2% between a first and second trial against 0.2% afterwards. Your first ROUVY test is the least reliable one you will ride, and unfamiliarity pushes it down.
Last, the day. ROUVY's ramp ends when you can no longer hold the target, and where that point falls is set as much by the previous five days as the previous five minutes: a heavy week, a warm room, short sleep, thin fuelling. Then decide. Under about 2%, with the same source and protocol, keep your old number. Between 2 and 5%, on a rested day and a clean pairing, accept it. Over 5%, accept nothing, average nothing, and repeat after two easy days.
Confirming a lower number is a bigger act than the button makes it look, which is the argument our pillar on training with ROUVY makes about the platform generally. ROUVY decides how a session feels, with real precision and a deep library behind it. Something upstream still decides what the session should be, and here much of that lives in the FTP field: press confirm and the library steps down with you [ROUVY Workout Library] while the same rides score more TSS [ROUVY TSS and IF].
AdaptCycling reads your ROUVY rides, including the recorded file, and fits a power curve to the efforts inside them. We do not send workouts to ROUVY. That curve can carry your threshold between tests and cut how often you need one, which our guide on FTP tests versus the power curve argues at length. It is not a replacement for testing: the estimate needs a starting number and proposes one upward when your rides earn it. A suspected decline is exactly where a ROUVY ramp earns its 40 minutes.
Quick answers
Which ROUVY FTP test should I take?
Why does ROUVY ask for my FTP before a 20-minute test?
My ROUVY FTP came back lower than last time. Have I lost fitness?
Should ERG mode be on for a ROUVY FTP test?
Sources cited in this guide
- 01
- 02
- 03
- 04
- 05
- 06
- 07Bouillod et al. 2017. Validity, Sensitivity, Reproducibility, and Robustness of the PowerTap, Stages, and Garmin Vector Power Meters in Comparison With the SRM Device. International Journal of Sports Physiology and Performance.
More inside Training with ROUVY
Start here · Foundational guide
Training with ROUVY: a self-coached cyclist's guide
ROUVY runs structured workouts properly and decides almost nothing. What its plans cover, where they stop, and how to train around them.
Read the full guide
Other articles in this series
- 01
Are ROUVY's training plans any good?
Coach-written, four to six weeks, included in the subscription. What they do well, and the one thing they do not do after a bad week.
- 02
Training on ROUVY without a power meter
ROUVY scores training load best from power. How to train and progress on heart rate alone, and what the ledger loses when you do.
- 03
How to import a workout into ROUVY
ROUVY accepts .zwo, .erg and .mrc and nothing else. The three ways a session gets in, and why the file sometimes lands wrong.
- 04
ROUVY ERG mode on routes: what the gradient does
Inside a structured workout with ERG off, ROUVY simulates a flat 1% and the filmed climb stops setting resistance. Why that surprises riders.
- 05
Zwift acquired ROUVY: what changes for riders
Zwift bought ROUVY and FulGaz in April 2026. What both companies committed to, what has actually changed since, and what has not.
Free training analysis · No card · ~3 minutes
Try the adaptive coach yourself.
Connect your rides and the coach reads your last eight weeks — intensity balance, ramp rate, what's working, what's holding you back — then drafts your training plan around it.
Free 14-day trial, no card.