20 Minute FTP Test protocol

Hi everyone and coach @chad,

I was looking at the 20 minutes FTP test protocol on Zwift and the original book by Andy Coggan. It seems as though the resting period between the 5-minute all-out and the actual 20-minute effort is 10 minutes rather than the 5 we have in the Trainerroad software. I suspect this can affect the results since after the all-out warm up we are pretty fatigued. Would you be able to elaborate on it?

P.s: Tagging also @IvyAudrain.



1 Like

To determine FTP, TR recommends using AIFTP detection as the preferred method, then the ramp test, and probably a 20 minute test or other method lastly. I don’t believe TR is going to revisit the efficacy of the 20 min test.

Hey, Amir!
@MI-XC is right that AI FTP Detection is the best option for you here. We created this tool to help address the shortcomings that come along with any singular FTP test from an isolated ride. AI FTP detection will look at a much broader range of rides and workouts, and will give you a much more accurate result free of any pacing or effort-gauging guesswork.

In fact, AI FTP Detection is 38% less likely to overestimate your FTP vs. the 20-Minute Test, and 75% less likely to underestimate your FTP vs. the 20-Minute Test.

More info here! ‘FTP Testing is a Thing of the Past: Introducing AI FTP Detection’.


I believe you and @IvyAudrain misunderstood my question.

I didn’t ask whether or not an AI FTP test is better than a Ramp or a 20 Minute test, my question was the reason why the 20 min FTP test protocol in TR is different than the original method mentioned by Coggan et al.


  • Amir

I understood the question completely, but you’re looking for semantic answers to an antiquated and no longer recommended protocol that probably meant little to nothing even at the time.

We could discuss the pros and cons of why one would recommend delivering DC versus AC electricity or you can accept AC is proven better now and move on.

There are plenty of more important things to worry about in cycling, like what zone 2/VT1/endurance training is :roll_eyes:, so just use AIFTP and get faster!


Agreed, use the AIFTP, but your question, I’m sure this will have already been debated to death in the forums over the many years, some searching and reading of the various TR threads and I’m sure you’ll find plenty of differing opinions of various testing protocols and there differences/rest periods for similar ones, why TR were different etc.

You could then maybe report back to the thread with your findings, some bullet points?

LOL, guy has a question and gets told to look it up himself and write a book report.


Recteating these protocols in an online training tool is pretty sub optimal. Z and TR both prescribe a 5 minute clearing effort at 110% - quite far from all out for most people surely?!

I’d prefer the shorter 5 minute recovery interval for a more functional FTP proxy, as 5 mins would give the anaerobic battery less time to recharge, making it more of an aerobic effort than with the 10 minute rest.

@AmirJamez - if you want to, you can recreate the 20 minute test to your specifications using trainingpeaks or trainerroad’s workout creator.

1 Like

But then surely the response to that is “Why have the 20 min test at all then?”. It appears TR hasn’t entirely ‘moved on’.

Perhaps AmirJamez’ question could be reframed as “If you’re going to offer a 20 minute test, why do it this way instead of the other way?”

1 Like
  • Keeping a workout in the library (for use by anyone that chooses it) is not the same as what they recommend and apply to people’s actual training calendar.

  • Purging a workout (which I believe happens rarely in TR, if at all) would seemingly be reserved for removing something considered harmful or undesired to a large degree.

  • Yes, TR believes in the new AIFTPD and they still prescribe the Ramp Test in those places where AIFTPD is unavailable (mostly new users). But even with that emphasis, they recognize that people may still prefer the 1x20-min or 2x8-min tests and keep them in the library to facilitate that (vs forcing people to make their own or go elsewhere).

  • As far as keeping the 1x20-min at the current layout, I presume that would be done for the sake of consistency with the history that they had with the test over time.

  • It may well be interesting to know why TR adopted the layout they did, but aside from the academic aspect I don’t seem much to be gained here for TR. As with any test protocol implementation, one of the key aspects is to maintain consistency from test to test. That assumes one reason for the test is comparison over time.

  • Comparing between different protocols, even ones that are nearly identical with a noted difference here, is not ideal and should be avoided. In short, “Pick a test and stick with it.” per the old TR suggestions is key here.

  • Maybe a TR rep can re-dive into that history to find the reasoning behind their implementation, but I see minimal benefit or need to do so.


But nobody said it was.

And I think this is the only bit that’s actually relevant to what was originally asked.

I don’t have a dog in that fight. I don’t think the OP did either. I don’t think there even was a fight here until someone waded in to answer a different question to what was asked, and tell the OP not to ask the question they did. My only point is that that doesn’t seem the most helpful or productive approach. :person_shrugging:


No fighting from me either, just sharing some thoughts / perspective.

Ultimately, the ball is in TR’s court again, if they choose to answer the OP core issue.



Got it!
In the question of: why does the TrainerRoad 20-minute FTP test have a 5-minute rest after warmup/prior to test, when the Andy Coggan model has a 10-min rest after warmup/prior to test?
The TrainerRoad 20-minute test isn’t intended to emulate the Andy Coggan model directly, we don’t claim to.

If you want more rest after your warmup prior to testing, you are welcome to pause for an additional 5 minutes.

Let me know if you have any further questions! :slightly_smiling_face: