AI FTP Detection Is Now Available for Everyone! ๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰


  • AI FTP Detection is available to all TrainerRoad subscribers!
  • AI FTP Detection is 38% less likely to overestimate your FTP vs. the 20-Minute Test
  • AI FTP Detection is 75% less likely to underestimate your FTP vs. the 20-Minute Test
  • AI FTP Detection is 40% less likely to underestimate your FTP vs. the Ramp Test
  • Powered by a machine learning model tested and improved through 1,200 iterations
  • Validated with over 22,000 athletes over 9 months
  • Takes into account indoor and outdoor rides
  • No need to do an โ€œall-outโ€ effort


Iโ€™m proud to announce that AI FTP Detection is now available to all TrainerRoad athletes!

For those unfamiliar with AI FTP Detection, you simply click a button to get an FTP! Our AI detects your FTP based on your indoor and outdoor rides when you click the button!

The cool part is that you donโ€™t have to do an โ€œall-outโ€ effort to get your FTP. You could even only do endurance work and still have AI FTP Detection predict your FTP.

And then AI FTP Detection, combined with Progression Levels, sets you up for productive training that makes you faster!

It took the team years to get to this point. In fact, we tested over 1,200 iterations of our ML model. This really was a huge effort!

Weโ€™re able to do this because of our massive 150,000,000+ ride database.

Over the last nine months, weโ€™ve had over 22,000 athletes sign up for AI FTP Detection. We used this time to find bugs, fix them, and validate our ML (Machine Learning) model(s).

Weโ€™re to a point where we can confidently say AI FTP Detection is ready for use, and is the recommended way to get your FTP on TrainerRoad! :tada:

I want to thank everyone on the forum whoโ€™s helped make this happen! Yโ€™all helped us test many unique cases and put AI FTP Detection through the wringer! Thank you!!

Weโ€™re not stopping here, either! Weโ€™re already working on the next batch of improvements to this feature based on the feedback youโ€™ve given on the forum.

If you want to learn more, I wrote a detailed blog post about how we got here and what AI FTP Detection means to TrainerRoad and cycling. You can read that here: FTP Testing is a Thing of the Past: Introducing AI FTP Detection - TrainerRoad Blog

If youโ€™ve found AI FTP Detection helpful, the team would love to hear that! Please post on this thread! Iโ€™ll collect them and then post them internally so the entire team can see the outcome of their hard work!!

We also know forum people like to ask questions, so here are some FAQs. If you have any other questions, please post them below.


AI FTP Detection FAQ

How is this different from FTP detections on other platforms?

Most other platforms estimate FTP from individual best efforts in your training data. Our model analyzes your training history and personal biometrics for a broader, more nuanced understanding of your abilities than a single effort can reveal.

Is AI FTP Detection just looking at what athletes generally gain in FTP after each block and basing the model on that?

No, we are not estimating FTP based on any static equation. AI FTP Detection uses a model trained on our data set to predict your FTP based on your unique training and biometric data.

Does AI FTP Detection consider rides that arenโ€™t TrainerRoad workouts?

Yes. AI FTP Detection takes into account all rides with power dataโ€”indoor or outdoor, structured or unstructured. This can include group rides, races, and even rides on other indoor training apps synced to TrainerRoad.

Will AI FTP Detection consider unstructured outside rides as well?

Yes. AI FTP Detection takes into account all rides with power data, whether theyโ€™re indoors or outside, structured or unstructured. This can include group rides, races, and even rides on other indoor training apps synced to TrainerRoad.

What does AI FTP Detection look at to detect my FTP? What types of efforts or workouts does AI FTP Detection consider?

AI FTP Detection looks at your training history and your biometrics to detect your FTP. Because AI FTP Detection considers so much of your training data, itโ€™s very unlikely a single type of effort or workout would substantially sway the model. For example, the model โ€œknowsโ€ when youโ€™re doing an easy ride and wonโ€™t confuse that for decreased fitness. You also donโ€™t have to do capacitive or all-out efforts for the feature to work. In fact, it works if you only do sweet spot or aerobic riding and never do any high-intensity intervals!

Will AI FTP Detection detect my FTP if Iโ€™ve never used TrainerRoad?

AI FTP Detection requires 10 completed indoor TrainerRoad workouts to begin detecting your FTP. Once you reach this minimum, you donโ€™t need to do any additional TrainerRoad workouts. By enabling RideSync, you can import rides with power data from other services and continue detecting your FTP.

Does TrainerRoad recommend I use AI FTP Detection or do an FTP test?

AI FTP Detection delivers the training benchmarks you need to progress your fitness and allows you to complete a productive workout in place of testing. Because AI FTP Detection gives you productive training stress without the unnecessary psychological and physical stress of an FTP test, we recommend using AI FTP Detection. That said, you always have the option to test!

Can I still complete an FTP Test?

Of course! You can still take a Ramp Test or other FTP test whenever you like. We just donโ€™t know why youโ€™d want to. :wink:

How often will AI FTP Detection give me a new FTP?

Depending on your current fitness, FTP generally takes several weeks to meaningfully change, so AI FTP Detection generally wonโ€™t detect a new FTP more often than this. For now, this is why AI FTP Detection isnโ€™t available more often than every 28 days.

What if my Abilities Change Within 28 Days?

FTP is not the only measure of your fitness. Between AI FTP Detections, your Progression Levels monitor your progress and track subtle changes in your abilities from day to day, even when your FTP doesnโ€™t meaningfully change.

Progression Levels track your capacity to express your current FTP in different training zones at any given time. For example, increasing your ability to hold steady power within your VO2 Max training zone from one minute to three minutes at the same FTP demonstrates significant progress and is reflected by changes in your Progression Levels.

Simply put, your abilities will change between AI FTP Detections in ways FTP alone canโ€™t accurately reflect. Progression Levels have you covered, working hand-in-hand with AI FTP Detection to track every change in your fitness, whether big or small, up or down.

I Didnโ€™t Accept my recent AI FTP Detection. Can I Get a New Detection Sooner Than 28 Days?

If you keep your current FTP instead of accepting AI FTP Detectionโ€™s results, the newly-detected FTP will still be available for you to accept at any time during the next 28 days. Simply click the AI FTP Button from your Account Settings or from any FTP Test, and youโ€™ll again be presented with the option to accept it, along with a countdown to when your next detection will be available.

If you havenโ€™t accepted your newly detected FTP and something happens to significantly decrease your fitness within 28 days (for example, if you take extended time off the bike), AI FTP Detection will automatically adjust your still-available prediction downwards to make sure you get the right workouts and power targets. This will prevent you from accepting a too-high FTP, and it wonโ€™t affect the 28-day countdown until your next detection is available.

Does AI FTP Detection predict my hour power or expected Ramp Test result?

AI FTP Detection isnโ€™t meant to predict the results of any individual effort, because all-out efforts can be affected by nutrition, fatigue, and other subjective factors. Instead, itโ€™s designed to give you a training benchmark (FTP) that represents the full scope of your abilities, so you get the most productive training possible.

Isnโ€™t AI FTP Detection just making training easier, and is that not the point of training?

A typical FTP test requires an all-out effort, but due to short durations and high resulting fatigue, these tests donโ€™t generally contribute as much useful training stress as a productive workout. AI FTP Detection lets you replace your FTP test with a beneficial workout and avoid a stressful, less productive effort.

How can AI FTP Detection be accurate if Iโ€™m coming back from an offseason or more than one week off the bike?

AI FTP Detection was trained on an enormous data set and can predict decreases in FTP even after long (weeks to months) training interruptions.

How Do I Know AI FTP Detection is Accurate?

The machine learning model powering AI FTP Detection was trained on a dataset with more than 150 million workouts and hundreds of thousands of FTP changes, including athletes of all different ages, abilities, and experience levels. But the best way to validate AI FTP Detectionโ€™s accuracy is to try it for yourself! Accept your detected FTP and try some productive workouts. We think youโ€™ll be amazed at the results.



Ok Iโ€™ll ask the obvious question:

What about overestimate vs ramp test? Any data vs 8-min test?



Iโ€™ve found it so simple to use I forgot it was in beta! I saw this post and thought duh. Then I realised itโ€™s now officially released lolz.
Great idea, easy to use and so glad I donโ€™t do ramp tests anymore.


Congrats to the TR team on getting this out.

Iโ€™ve a question on L:R power balance and how this impacts the AI FTP.

I use a Neo turbo indoors and during the drier months, I use Assiomas outdoors. I donโ€™t really want to keep replacing Assioma bearings throughout the British winter and want to use a simple left-hand crank meter, probably 4iiii. However, my Assiomas indicate a 44:56 power balance.

If I swap around power measurement devices Iโ€™m guessing with my power imbalance that this may impact the usefulness of the AI FTP?

I know I can use scaling on the left-hand crank, but what should I set this to? I donโ€™t know whether the balance is closer to 50:50 at higher interval wattages vs lower power outdoor cruising around rides.

There has been plenty of discussion about TR not visualising L:R balance or even allowing it to pass through to other ride sync sources. I think the rationale is how does it make the rider faster to have this data. Could there now be a case for recording L:R balance and even for the AI FTP data to prescribe the scaling factor when using different power measurement devices to assist AI FTP usefulness?

Cheers Scott

Well done TR. So glad FTP AI is now automatic.
20 mins Ramp tests are a crazy VO2 test in my opinion. Highest score I ever got was 242. Though I have managed 290 NP for 2 hours on Barren. Therefore, I have managed my FTP manually. But today TR AI has given me a 300 FTP !!! Happy days. 5 above my manual prediction


Iโ€™d forgotten this was in Beta to be honest, great news.

Iโ€™m a little confused as the FAQ article indicates:

  • Indoor workouts with power
  • Indoor workouts with only an HR monitor
  • Outdoor workouts with power
  • Outdoor workouts with only an HR monitor

But the above abridged version suggests only power is used.

Can someone confirm which is correct?


Nice :+1: now for the next step, are there any plans to recommend certain types of workouts and schedule to an athlete which would maximise their FTP growth (instead of the fixed training blocks)?


AIFTP is the best thing since sliced bread (or actually โ€œalternatesโ€ in the TR-world :grinning: ) Iโ€™ve always struggled with the Ramp Test, especially at the beginning of the plan when switch to indoor training. Workouts have always felt too easy. E.g. I never understood the fuzz about O/Us being so hard. This season AIFTP has given me an ATH FTP for season start. Ego feels good, but the legs hurt :crazy_face: And now I finally DO understand the fuzz about O/Usโ€ฆ

My only concern is similar to what was posted above, namely multiple power sources. I use a Kickr Core inside, and have a dual sided 4iiii on my road bike. They work pretty well together. But I am reluctant to drop any money on a power meter for my mtb. Different bike, different meter, different type of riding. So at least for me Iโ€™ll be riding the mtb by RPE. Saves me some $$$, but I do miss some structure and numbersโ€ฆ

I made the move from Zwift training to TR in June and it has to be said TR is working really well for me. The AIFTP and AT pairing seems to work just perfectly for me. Iโ€™ve managed to go from 226w to 263w according to AIFTP.

The thing is I donโ€™t know if those numbers are โ€˜correctโ€™ but what I can tell you I do feel over 15% stronger and I do think that the training is aimed correctly at meโ€ฆ all workouts seem to โ€˜do the jobโ€™

Keep up the good work guys.


Huge congrats to the TR team! I sincerely think this will be a massive game changer, not just within TR, but for power based training across all systems.

Now that itโ€™s live, how can we remove all the ramp tests from our training plan, and have them replaced with the optimal workouts to compliment the rest of the plan and continue progression most effectively? Seems quite a heavy restructure & reshuffle of the entire TSS build within the Ramp Test weeks would be called for, possibly switching the other workouts and focuses in those weeks entirely.

It seems at first thought that a lot more progress per month could now be made both within those weeks, and in your plan overall, now that these โ€œcompletely fry yourselfโ€ efforts every 4 - 6 weeks have been removed. Iโ€™m guessing fully capitalizing on that will need pretty deep plan restructuring, at least within those weeks, not just manually swapping the Ramp Test for a randomly chosen workout.

Particularly for riders partway through a 6 - 12 mo plan, โ€˜Delete and build a new plan.โ€™ may not be a great option.

Apologies if I missed an already existing answer and/or How To, but had a real search both on site and forums, and couldnโ€™t find it!

Thank you for an AMAZING quantum leap forward. Canโ€™t wait to see where it takes people!

1 Like
  • I donโ€™t think they have a setup that just swaps them all out in one step.

  • Per the defined process, you run the AIFTPD tool by clicking the button located next to your Ramp Test on the calendar (as applied via plan addition or manually adding a RT). You run that detection, accept the result and will be offered a replacement workout if you choose.

  • Essentially, you just do that detection & swap as you get to that point in your calendar, not ahead of time.

This is a tricky one. Youโ€™re basically saying that one of your power meters doesnโ€™t measure power the same way as your other power meter.

Thereโ€™s a sideway into this. WLV2 takes into account indoor and outdoor power and might do an offset depending on your historical trends. So if youโ€™re always lower/higher on one, then it should take that into accountโ€ฆif itโ€™s consistent.

The team has an issue to do not just splitting indoor/outdoor but do it per power meter. But thatโ€™s a little lower on the priority list since not many people ride consistently with three power sources.


Rides with just HR are considered, but itโ€™s always better to have power on all of your rides. Itโ€™s easier for the ML to figure out your FTP if you have power on all of your rides.


Weโ€™re always working on optimizing training. In our next release, weโ€™re going to take people to different progression levels depending on your plan.

The current plan is, if youโ€™re in specialty, weโ€™ll let you train up to level 9-10 in your primary zones. Then if you hit that, weโ€™re just going to bump you up 2% so that you donโ€™t get a drastic FTP increase and start lowering your TIZ or how many repeats youโ€™re doing.

IE if youโ€™re in week of 6 of 40k TT you donโ€™t want to keep practicing your sustained power at really long durations rather than building up from a new FTP.


Yes, we have two phases for this:

Current release: If you have a ramp test within 28 days of your last AI FTP Detection it will be replaced with a workout.

Future release: AI FTP Detection goes full auto, and youโ€™ll have the choice not to have tests on your plan. After 28 days, weโ€™ll automatically run AI FTP Detection after every workout and let you know when it increases in a meaningful manner.


@Nate_Pearson Aaaaamazing!! You guys ROCK my CASBAH!!!


Had originally posted this in the other AI FTP thread and the thread got locked like an hour later :confused:

"Looking to add some extra โ€˜volumeโ€™ to my low volume traditional base plan I am doing over the winter (in week 2 of Trad Base I currently) and would like to choose the option that will have the most impact on my AI FTP. Which of the following options would help boost my AI FTP calculation?

A) choosing a higher PL workout as an alternate once or twice per week (Klammspitze instead of Fletcher, 20 more TSS, extra 30 mins, 0.1 lower IF, 0.4 higher PL)

B) adding an extra lower level endurance workout once per week (For example Carter, 3.3 endurance, 34 TSS, 45mins)

C) adding an extra lower level threshold workout once per week (For example Sanford -1, 1.5 threshold, 50 TSS, 45 mins)

Is the answer based on boosting your TSS the most or your PL? I can only do so much but with the extra work I put in whether it is the same amount of workouts but harder or an extra workout, I want it to have the most impact on my FTP.

I expect both are factors but at the start of this plan AI FTP suggested a drop (I rejected it) put in a ticket and they couldnโ€™t tell me why a drop was suggested but have continued on and havenโ€™t had an issue.

If someone like Ivy looks into things related to that, I also had an issue last week where I did my Sunday workout (Gibraltar -3) on Saturday and it disappeared from Sunday on the App but stayed on the Sunday on the web calendar."


Sounds like youโ€™re saying Iโ€™d be wasting my money buying the left only 4iiii. Just going HR only seems a shame after all your teams hard work. Iโ€™ll have a think about what the best vfm, wet weather total power meter might be. It could be use the Assiomas and accept the bearings might need a couple of replacements over the wet months.

I would use a left-sided only PM over a HR monitor, even if your situation.

If your offset is consistent, then you could use the offset feature in 4iiii software


Congrats to Nate and the TR team! You all should be very proud of your hard work. This ecosystem of AI FTP detection, Progression Levels, Adaptive Training, TrainNow, and the podcasts makes TrainerRoad such an amazing training platform. I love it. Especially TrainNow which has become my main way to train.

I was in the beta and can say that AI FTP detection was accurate for me whether going upwards or downwards, the subsequently prescribed workouts at the new ftp just felt right and productive.

Thank you!