How To Calculate Your FTP: Tests, Zones And What's A Good Number
Share
Your FTP is the number every training zone you own is built on, so getting it right matters a great deal more than getting it high. It is the highest power you could hold for about an hour. The quickest honest way to find it is a 20 minute effort multiplied by 0.95, and if you have ridden hard in the last 90 days you do not need to do a test at all, because the number is already sitting in your ride files.
Key takeaways
- FTP is the highest power you could hold for about an hour. It is the number your entire training plan is built on.
- To work it out from a 20 minute effort, take your average power and multiply by 0.95. Held 250 W for 20 minutes? Your FTP is around 238 W.
- You do not have to do a test. If you have ridden hard in the last 90 days, the number is already in your ride files. Find your FTP from your existing rides here.
- Ramp tests tend to read high, especially for strong, muscular riders, because they reward a type of fitness that has little to do with your threshold.
- The aim is the most accurate number, not the highest one. An inflated FTP means every session you do afterwards is too hard to hold.
- Recheck it every 8 to 12 weeks, or after any block that has clearly moved you on.
What is FTP in cycling?
FTP stands for Functional Threshold Power. It is the highest average power, in watts, that you could sustain for roughly an hour before fatigue forces you to ease off.
It matters because almost nothing else in your training works without it. Your zones come from it. Your interval targets come from it. Whether a session was easy, steady or genuinely hard is judged against it. Get the number right and everything downstream is right. Get it wrong and every ride you do is slightly the wrong intensity, week after week, for months.
That is why I keep saying the same thing to riders: step one is knowing your zones, step two is training in them. Most people skip straight to step two and wonder why the plan is not working.
How do you calculate FTP?
The standard method is to ride as hard as you can hold for 20 minutes, take your average power for that effort, and multiply it by 0.95.
Worked example: you average 250 W for 20 minutes. 250 × 0.95 = 237.5, so your FTP is about 238 W.
The 5% reduction is there because 20 minutes flat out is a harder effort than an hour flat out. Nobody sensible wants to ride a maximal hour just to get a number, so the 20 minute test with the adjustment became the practical standard.
If you ever do ride a genuinely maximal 40 or 60 minute effort, use that average directly with no adjustment. A real hour is the most honest reading you will ever get.
The four ways to find your FTP, compared
| Method | How long | What it is good at | What it gets wrong |
|---|---|---|---|
| 20 minute test | 20 min plus a long warm up | The best balance of accuracy and practicality. The number most coaches, including me, work from. | Needs proper pacing. Go out too hard and you will fade badly and under-read your own fitness. |
| Ramp test | 8 to 25 min | Quick, easy to execute, hard to pace badly. Popular on Zwift and smart trainer apps. | Tends to read high. It rewards anaerobic power and raw strength. Riders with gym legs and limited aerobic fitness can post big ramp numbers they cannot hold for an hour. |
| 8 minute test | 2 × 8 min | Less unpleasant than 20 minutes. Useful for riders who dread a long effort. | Even shorter, so even more anaerobic contribution. Usually needs a bigger correction and is less reliable. |
| One hour time trial | 60 min | The real thing. No estimate, no correction factor, no argument. | Brutal, hard to pace, and hard to find a road where you will not be interrupted. Almost nobody does it twice. |
| From your existing ride data | No test at all | Uses efforts you have already done. Nothing to schedule, nothing to dread. | Only as good as the hardest effort in your files. If you have not been hard in three months, it will read low, and it should tell you so. |
Can you find your FTP without doing a test?
Yes, and for a lot of riders it is the better option.
Your FTP is already hidden in your ride history. If you have done a hard group ride, a hilly loop, a club 10 or any effort where you properly emptied yourself in the last 90 days, the data to work it out is sitting in your files right now. You do not need to schedule anything or dread anything.
I built a free calculator that does exactly this. You drop in four or five of your hardest recent rides, and it finds your best 20 minute power, works out your FTP, pulls your threshold heart rate from the same effort, and gives you your full set of training zones. It runs entirely in your browser, so your ride files never leave your computer.
Find your FTP from your existing rides
It will also tell you something most calculators will not: whether it thinks that effort was genuinely flat out. If your shorter efforts suggest you had more in you, it says so rather than handing you a number that is quietly too low.
What about Strava's estimated FTP?
If you pay for Strava, go to Training, then Power Curve, set a custom date range covering the last 90 days, and switch on "Show Estimated FTP". It will overlay its own estimate.
It is a reasonable starting point and it costs you nothing extra if you are already subscribing. Be aware that plenty of riders find Strava's estimate runs high, for much the same reason ramp tests do: short, punchy efforts pull the estimate up. Worth checking it against a number worked out from a real sustained effort before you build a training block on it.
If you are on free Strava, you do not get the Power Curve at all. You can still export your own rides, which is what the calculator above is built for.
What is a good FTP for a cyclist?
Raw watts on their own tell you very little. A 280 W FTP is strong on a 60 kg rider and fairly ordinary on a 95 kg one. What matters is watts per kilogram: your FTP divided by your body weight in kilos.
| What it usually means you can do | Men (W/kg) | Women (W/kg) |
|---|---|---|
| Getting started | under 2.0 | under 1.6 |
| Riding regularly, building the base | 2.0 to 2.8 | 1.6 to 2.3 |
| Comfortable on long, hilly days | 2.8 to 3.5 | 2.3 to 2.9 |
| Strong on the climbs, near the front of the group | 3.5 to 4.2 | 2.9 to 3.5 |
| Racing, and competitive locally | 4.2 to 4.9 | 3.5 to 4.3 |
| Racing at a high amateur level | 4.9 to 5.4 | 4.3 to 4.8 |
| Domestic professional | 5.4 to 6.0 | 4.8 to 5.3 |
| International professional | above 6.0 | above 5.3 |
The top three rows are anchored to Andrew Coggan's power profile chart, which is still the standard reference for what a professional actually produces. The rest I have described by what the number tends to let you do on a bike, because that is what people are really asking.
I have deliberately not used the words "untrained", "fair" or "moderate", which is how the original chart labels the bottom rows. I coach a rider at 2.6 W/kg who rode the Struggle Tour of Britain. Another at 2.5 who completed a hundred mile day in the Cotswolds. A third who got round Etape Loch Ness after heart surgery, on medication that caps his heart rate. Filing those three under "non-racer" tells them nothing true about what they can do.
Watts per kilo does not paint the whole picture
Here is the bit almost nobody says out loud, and it matters more than the table above.
Your FTP describes one hour. That is all it describes. It tells you nothing at all about what happens in hour four, or hour seven, or hour nine.
Two riders can have exactly the same watts per kilo. One holds a decent percentage of it all day and finishes a mountain sportive strong. The other is in real trouble by lunchtime. Same number on the chart, completely different day. The difference is durability, and it is the thing that actually decides how a Mallorca 312 or an Etape goes.
I am a good example of this myself. My power to weight is nothing special and I would not trouble the upper rows of that table. But I can hold what I have got for a very long time, and on a twelve hour day that counts for far more than another half a watt per kilo would. I have finished ahead of riders with better numbers than mine, plenty of times, because the number was only ever describing their best hour.
The honest position is that durability is hard to measure well. There is no neat single figure for it the way there is for FTP, and nothing on a chart captures it. What I look at with my own riders is whether their hard efforts hold up late in a long ride, once they have several thousand kilojoules in the legs. If your five minute power after four hours looks like your five minute power when fresh, you have got something the chart cannot see.
So take the table as orientation and then get on with it. If you ride hilly sportives or something like the 312, watts per kilo matters on the climbs but durability decides your day. If you ride flat time trials, absolute watts matter more than either. Pick the measure that suits what you actually do, and remember that the only number that means anything is your own, measured the same way, three months apart.
Why accuracy matters more than a big number
This is the part people get wrong, and it is the reason I am fussy about it.
An inflated FTP does not make you fitter. It makes every session you do afterwards too hard. Your sweet spot work becomes threshold work. Your threshold intervals become VO2 efforts you cannot complete. Your easy rides creep up because the zones underneath have all shifted. You end up permanently slightly too tired, never quite recovering, and unable to work out why a sensible plan is not producing anything.
An FTP that reads low is a gentler problem but still a real one. You train too easy, you never touch the intensities that would move you on, and you plateau while doing everything you were told to do.
I see inflated numbers far more often than low ones, and ramp tests are usually where they come from. A rider with a strong, muscular build and a decent anaerobic kick can produce a ramp number that looks excellent and bears very little relationship to what they can hold for an hour. It is not their fault and it is not cheating. The protocol simply measures something adjacent to the thing you actually want to know.
None of this is about ego. It is about the number being right, because everything else is built on top of it.
How to get the most out of an FTP test
If you are going to do a formal test, the four things that change the result most are: warm up properly for at least twenty minutes, do not go out too hard in the first two minutes, get a fan on you if you are indoors, and zero-offset your power meter before you start.
I have written a whole piece on getting a test right, because there is more to it than most people think and the difference between a well executed test and a scrappy one is easily twenty watts.
Read: how to maximise your FTP test
Once you have the number, train with it
Knowing your threshold is step one. Step two is doing the work at the right intensity. I have put three of the sessions I use with my own coaching clients up as free downloads, ready to drop straight into Zwift or TrainingPeaks, and every one of them is written as a percentage of your FTP so they scale to whatever number you just found.
Download three free sessions to raise your FTP
How often should you retest?
Every 8 to 12 weeks in normal training, or sooner if a block has clearly changed things. If sessions that used to hurt now feel manageable, your number has moved and your zones are stale.
You do not always need a formal test to check. Run your recent rides back through the calculator and see whether your best 20 minute power has crept up. If you race, or ride a club 10 mile time trial regularly, that effort is effectively a rolling FTP check without you having to schedule anything.
Frequently asked questions
What does FTP stand for?
Functional Threshold Power. The highest average power in watts you could hold for around an hour.
How do I calculate FTP from a 20 minute test?
Take your average power for the 20 minutes and multiply it by 0.95. So 250 W becomes roughly 238 W.
Can I find my FTP without doing a test?
Yes. If you have ridden hard in the last 90 days, your ride files already contain the effort needed. Our free calculator works it out from your existing rides, in your browser, without uploading anything.
Is Strava's estimated FTP accurate?
It is a reasonable starting point, but it is only available on a paid Strava subscription and many riders find it reads high, because short hard efforts pull the estimate upwards. Cross-check it against a sustained effort.
Why is my ramp test FTP higher than my 20 minute FTP?
Because a ramp test rewards anaerobic power and muscular strength alongside aerobic fitness. If you are strong but aerobically underdeveloped, a ramp will flatter you. The 20 minute number is usually closer to what you can genuinely hold.
What is a good FTP for a cyclist?
It depends far more on your weight than your watts. Divide your FTP by your body weight in kilos. Most trained club riders sit somewhere between 2.8 and 4.2 W/kg. Your own trend over time tells you more than any chart.
How often should I test my FTP?
Every 8 to 12 weeks, or whenever your sessions start feeling noticeably easier than they should.
Do I need a power meter to find my threshold?
No. You can work out your threshold heart rate instead, from the average heart rate during a hard 20 minute effort. Heart rate is less precise and it lags behind what your legs are doing by two to three minutes, so it is best used for steady riding rather than short intervals. The calculator will give you both if your files have both.
Ready to find your number?
You do not need to book a lab test or dread a session all week. If you have been riding hard, the answer is already in your data.
Find your FTP without doing a test
Testing indoors gives you the most repeatable number, which is why the trainer you use matters. If you are still choosing one, my guide to which turbo trainer to buy covers how much power accuracy actually matters and what to spend.
And if you want someone to look at the number with you and build the training around it, that is what I do. Have a look at 1:1 coaching.
Coach James
