Komeak
EN
Request access

All articlesThresholds

FTP: what it is, and how to test it

Functional threshold power explained: what FTP measures, the 20-minute and ramp tests step by step, and what research says about how accurate they are.

FTP — Functional Threshold Power — is the highest power you can hold, in a more or less steady state, for about an hour. The definition is Andrew Coggan’s, who put it forward in the early 2000s as a marker any cyclist could measure alone, on the road, without a lab.

If you ride with a power meter, it is the number everything else hangs on. Your training zones are percentages of it (Power and heart-rate zones). Your training load is computed against it (Training load). A wrong FTP skews everything downstream: zones that are too easy or too hard, and loads that no longer say what your activities cost you.

What FTP measures, and what it doesn’t

Below your threshold, you can keep going for a long time: your body clears lactate as fast as it produces it. Above it, lactate builds up, and the effort has an end in sight. FTP tries to put that boundary in watts.

It is close to lab thresholds, but no substitute for them. A study of 23 trained cyclists compared FTP from a 20-minute test, the power held for 60 minutes, and the individual anaerobic threshold measured from blood lactate: on average, all three agreed. But from one rider to the next, the differences were too wide to treat them as interchangeable. Another team reached the same conclusion: FTP is a valid field marker, and how far it sits from the lactate threshold depends on the athlete’s fitness.

So take FTP for what it is: a practical marker you can measure alone and repeat, used to set your zones and your load. Not a physiological value accurate to the watt.

The one-hour test

The definition itself: ride as hard as you can for an hour, at an even pace, and take the average power. It is the reference, and also the hardest test to get right: an hour of road with no stops, perfect pacing, and race-day motivation. Few riders do it.

The 20-minute FTP test

The most widely used test, as described by Allen and Coggan:

  1. Warm up for 15 to 20 minutes.
  2. Ride a 5-minute all-out effort. It empties your anaerobic reserve: without it, the 20 minutes that follow would draw on it, and your FTP would come out too high.
  3. Recover for 10 minutes, pedaling easily.
  4. Ride 20 minutes as hard as you can, at an even pace — ideally on a steady climb or a flat road with no lights or junctions.
  5. Cool down for 10 to 15 minutes.

Your FTP is 95% of your average power over the 20 minutes. You can’t hold your 20-minute power for an hour, so 5% comes off. For example, 280 W held for 20 minutes gives an FTP of 266 W.

FTP 266 WWarm-up5 min flat outRecover20 min · 280 WCool-down
The 20-minute test, each block in the color of its power zone: 280 W held for 20 minutes gives an FTP of 266 W.

The test is reliable from one attempt to the next, and its power is closely tied to the power you hold for an hour — as a study found when trained cyclists repeated each test twice. The 95% factor is still an average: for some riders, hour power is a little higher or a little lower.

A bonus if you also ride with heart rate: your average heart rate over the 20 minutes gives an indication of your threshold heart rate (Threshold heart rate: finding it without a lab).

The classic mistake: starting too fast. A 20-minute test is mostly lost in its first five minutes. Start slightly below your target pace, and finish flat out.

The ramp test

A short test with no pacing to manage, popular on indoor trainers:

  1. Warm up for 10 to 15 minutes.
  2. Start at an easy power, and add 20 W every minute, seated, until you can no longer hold the target.
  3. Note the average power of your best completed one-minute step.

FTP is estimated at about 75% of that power. That factor is a convention, not a measured constant: no published study has validated it. It assumes a fixed ratio between your best minute and what you can hold for an hour, when that ratio depends on the kind of rider you are: a punchy rider with a big anaerobic capacity gets an estimate above their true FTP, a diesel gets one below.

FTP 315 WBest step · 420 WWarm-up+20 W a minuteCool-down
The ramp test: 20 W more every minute. The last completed step, 420 W, gives an estimated FTP of 315 W.

The ramp test also gives your maximum heart rate: the highest value reached before you stop.

Which test should you choose?

Test Effort Best for Keep in mind
One hour 60 min Riders used to time trials The reference, but rarely well paced
20 minutes 5 + 20 min Most riders The best trade-off; you need to pace yourself
Ramp To exhaustion, in one-minute steps Beginners, indoor trainers Easy to get right; less certain depending on your profile

Whichever you pick, always repeat the same test, in the same conditions: rested, on the same bike, with the same power meter, indoors or outdoors but not one then the other. It’s the change from one test to the next that tells you how you’re progressing.

These tests call for a maximal effort. If in doubt about your health, ask a doctor before doing one. None of them is compulsory: an FTP entered from memory is better than none, and can be corrected later.

When to test again

Your FTP changes with your training: it rises through a block of work and falls after a break. Test again when it has probably moved — at the start of a new training period, after time off, or when your activities feel easier than they should at the same intensity.

In Komeak

Your FTP is set in Profile, Cycling. You say whether you ride with a power meter, then enter your FTP. Coggan’s seven power zones are computed from it straight away, and you can adjust them by hand.

Cycling

FTP262 W
Threshold heart rate168 bpm

Power zones

Z1Z2Z3Z4Z5Z6Z7
20-minute testRamp test

Your past seasons

  1. Since 1 Jan 2025Current

    FTP 262 W · Threshold HR 168 bpm

  2. 2023 – 2024Proposed
    FTPYour efforts: 20 min at 261 W
    248 W
    Threshold HRFrom your device
    164 bpm
    Validate
  3. 2020 – 2021No profile

    Neither power nor heart rate

  • Tests are explained step by step. Under Evaluation tests: Coggan’s test (20 min) and the ramp test for your FTP, next to the ones that give your heart-rate thresholds. Each comes with its full protocol, and says what it determines.
  • You can plan a test for a given day, from the Cycling page or from an FTP Komeak estimated: the Thresholds card on the home screen reminds you, and keeps it, flagged overdue, until you have done it, rescheduled it or canceled it. Once it’s done, I’ve done it asks for the day of the test and your average power: Komeak computes your FTP from it — 95% of the 20 minutes, or 75% of the ramp’s best step —, records it as determined by a test, and computes your activities since that day again.
  • Komeak proposes a higher FTP when your activities show it. It estimates it at 95% of your best 20-minute effort since your FTP last changed, and proposes it as soon as it is 3% above yours — on the home screen, in the Form tab and on the activity it comes from. You can dismiss it: it only comes back at 5%. From 7%, it stays until you accept it: every load is off by then. Accepting opens a new version, without touching the past.
  • Your thresholds are dated. Edit corrects a typing mistake; New version opens a new period, and your past activities keep the FTP their load was computed with. A March activity stays computed with your March FTP.
  • Your history has its FTPs too. Drop the archive of the platform you used: Komeak proposes a profile for each year before yours, with the FTP from your bike computer, or the one your efforts show when they beat it by 3%. You validate or correct each, the latest first, and its activities are computed again with it. Measuring power without knowing your FTP? Komeak proposes the one of your last 90 days.
  • Your FTP shows wherever it matters: on an activity’s page, the FTP used for its calculation and its intensity as a share of FTP; a reference line on its power curve; and on the Form tab’s curve, a marker on the day it changed, which explains a jump in the curve.
Sources
  • Allen H., Coggan A., McGregor S., Training and Racing with a Power Meter, 3rd ed., VeloPress, 2019.
  • Borszcz F. K. et al., “Functional Threshold Power in Cyclists: Validity of the Concept and Physiological Responses”, International Journal of Sports Medicine, 39(10), 2018, pp. 737–742. doi:10.1055/s-0044-101546
  • Valenzuela P. L. et al., “Is the Functional Threshold Power a Valid Surrogate of the Lactate Threshold?”, International Journal of Sports Physiology and Performance, 13(10), 2018, pp. 1293–1298.
  • MacInnis M. J., Thomas A. C. Q., Phillips S. M., “The Reliability of 4-Minute and 20-Minute Time Trials and Their Relationships to Functional Threshold Power in Trained Cyclists”, International Journal of Sports Physiology and Performance, 14(1), 2019, pp. 38–45. doi:10.1123/ijspp.2018-0100
  • Friel J., The Cyclist’s Training Bible, 5th ed., VeloPress, 2018.
On the home pageZones built on your own thresholds

Your next peak starts here.

Komeak is in private beta. Tell us how you ride, and we'll let you know when your place is ready.

Request access