FTP Test: 20-Minute vs Ramp Test (Which One to Use and When)
Field note #263 · 2026-05-30 · 5 min read
Rapid answer
The ramp test is more accessible and paces itself automatically. The 20-minute test is more accurate for trained threshold riders who know their pacing well. Both produce FTP estimates within 3 to 5 percent of each other for most athletes.
FTP (Functional Threshold Power) is the highest average power you can sustain for approximately 60 minutes. Two test protocols are commonly used to estimate it without doing a full 60-minute all-out effort: the 20-minute test and the ramp test. Both work. They suit different athletes.
20-minute FTP test vs ramp test: key differences
The 20-minute test produces a direct power measurement that you multiply by 0.95 to get FTP. The ramp test drives you to exhaustion in one minute increments and takes 75 percent of your peak 1-minute power. The 20-minute test rewards good pacing; the ramp test is self-pacing and harder to botch.
| Factor | 20-minute test | Ramp test |
|---|---|---|
| Duration | 40 to 50 min total (with warmup) | 20 to 30 min total |
| Pacing required | Yes, high skill requirement | No, each step is fixed |
| Failure mode | Going out too hard, fading | Rare (it escalates to failure) |
| Best for | Experienced riders with pacing discipline | Beginners, riders who fade in long efforts |
| Formula | Avg power x 0.95 | Peak 1-min power x 0.75 |
| Accuracy for trained riders | Slightly higher | Slightly lower |
| Recovery needed after | Significant (1 to 2 easy days) | Moderate (1 easy day) |
How to do the 20-minute FTP test
The 20-minute test protocol: warm up for 10 minutes at moderate effort, complete a 5-minute hard but not maximal effort to open up the legs, recover for 5 minutes, then ride at the highest average power you can sustain for exactly 20 minutes. Multiply the 20-minute average power by 0.95.
Full protocol:
- Warm up: 10 to 15 minutes at Zone 2 power.
- Opener: 5 minutes at roughly 110 percent of expected FTP (hard but not maximal). This clears glycolytic pathways and gives a more accurate 20-minute result.
- Recovery: 5 minutes easy spinning.
- Test: 20 minutes all-out, as even as possible. The 0.95 correction assumes you can sustain the last minute as well as the first. If you fade significantly in the last 5 minutes, your pacing was off.
- Apply the formula: 20-minute average watts x 0.95 = estimated FTP.
Common pacing mistake: starting too hard. A 20-minute effort should feel like "this is hard but I can hold it" for the first 10 minutes, not "I am already struggling" at minute 4.
How to do the ramp test
The ramp test increases power by a fixed amount (usually 20 watts) every minute until you cannot maintain the target cadence. Take 75 percent of the peak 1-minute average power. This test paces itself because each step has a fixed target.
Full protocol:
- Warm up: 10 minutes easy.
- Start at a low power (around 50 percent of expected FTP).
- Increase by 20 watts every minute.
- Continue until you cannot complete the minute at the target power.
- Identify the highest 1-minute block you completed fully. That is your peak 1-minute power.
- Apply the formula: peak 1-minute power x 0.75 = estimated FTP.
Most cycling apps (Zwift, TrainerRoad, Wahoo) have built-in ramp test protocols that automate the step increases.
The ramp test also captures peak aerobic power (MAP), which is useful for structuring VO2 max intervals.
Which test is more accurate?
For most athletes, the two tests produce results within 3 to 5 watts of each other. The 20-minute test is slightly more accurate for athletes who train primarily at threshold intensities, because FTP is defined as a threshold effort, and the 20-minute test directly measures threshold performance.
The ramp test tends to produce slightly higher FTP estimates for athletes with a strong neuromuscular profile (good sprint and VO2 max relative to threshold). The 75 percent multiplier was calibrated for riders with a typical power curve.
If your ramp test FTP produces training zones that feel too easy in Zone 4 intervals, your neuromuscular fitness is outpacing your aerobic fitness and the 20-minute test will be more representative.
How often should you test FTP?
Every 4 to 8 weeks during a build phase, or whenever a new training block starts. Testing more frequently does not produce better data. Testing every 6 weeks is a common standard.
Do not test during a recovery week or following a hard training block. A fatigued test will underestimate FTP by 5 to 10 percent. Test at the start of a fresh week after at least 2 easy days.
Use the FTP calculator to enter your test result and get all 7 Coggan power zones instantly.
For a comparison of smart trainers by power accuracy, see best smart trainers for FTP testing. For FTP-based zone setup on a cycling computer, see the best cycling computers guide. Triathlon athletes should also set separate run and swim zones; see triathlon training zones for how each discipline's targets differ from bike zones.
Frequently asked questions
How often should I test my FTP?
Test every 4 to 8 weeks during a training build phase, or at the start of each new block. Testing more frequently than every 4 weeks does not produce more useful data because fitness changes require time to accumulate. Every 6 weeks is a common standard. Avoid testing during a recovery week or after back-to-back hard days. A fatigued test underestimates FTP by 5 to 10 percent, which drags all your training zones down and makes the data useless for structuring future sessions.
What is a good FTP for an amateur cyclist?
A good FTP for an amateur cyclist is roughly 2.5 to 3.5 watts per kilogram of body weight. A 75 kg rider in that range would have an FTP of 188 to 263 watts. Cat 4 club racers typically sit at 3.0 to 3.5 w/kg, while serious recreational riders who train consistently land around 2.5 to 3.2 w/kg. These are rough benchmarks, not targets. What matters is your own trend over time: a rising FTP at a consistent weight means your training is working.
Can I do an FTP test on a Zwift trainer?
Yes. Zwift has a built-in ramp test and a 20-minute FTP test, both of which work well on any smart trainer that can broadcast power. The ramp test is the more common choice in Zwift because it paces itself automatically and eliminates the risk of going out too hard. If your trainer is not a direct-drive model, accuracy depends on the calibration of your trainer's power estimate, so calibrate your trainer before testing and use the same setup every time you test for comparable results.
Should I do a warm-up before an FTP test?
Yes, always. The 20-minute test protocol includes 10 to 15 minutes of Zone 2 warm-up, a 5-minute hard opener at roughly 110 percent of expected FTP, and then 5 minutes of easy recovery before the test begins. Skipping the opener produces a lower result because the glycolytic pathways are not primed. The ramp test protocol also requires a 10-minute easy warm-up before the steps begin. Never go straight into a test cold.