# Trainer Road FTP vs …FTP

**URL:** <https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694>\
**Category:** Training\
**Tags:** ftp\
**Created:** [February 1, 2026, 6:11am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694 "2026-02-01T06:11:59Z")\
**Posts on this page:** 20\
**Page:** 3

<div class="post-metadata">

**Author:** ![gyrerider](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/gyrerider/32/132372_2.png) [@gyrerider](https://www.trainerroad.com/forum/u/gyrerider)\
**Post date:** [February 3, 2026, 8:20pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/41 "2026-02-03T20:20:03Z")

</div>

Ha, yeah.. ignorance is bliss. If only I had the will power to not look at it!

---

<div class="post-metadata">

**Author:** ![brem](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/brem/32/122969_2.png) [@brem](https://www.trainerroad.com/forum/u/brem)\
**Post date:** [February 4, 2026, 4:41am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/42 "2026-02-04T04:41:51Z")

</div>

The upgraded TR app is working brilliantly for me. Workout selections push hard but am able to complete them - some comfortably. Body is always adapting and improving to harder workouts, so long as I take yellow days off along with rest. Power outputs consistently increasing. Holding higher power for longer across all zones. FTP increased and is predicted to increase more. Functional power accords with actual power as best I can make out. I’m super impressed with the new TR upgrade - seems to be working flawlessly for me.

---

<div class="post-metadata">

**Author:** ![harkoo](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/harkoo/32/7746_2.png) [@harkoo](https://www.trainerroad.com/forum/u/harkoo)\
**Post date:** [February 4, 2026, 6:17pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/43 "2026-02-04T18:17:44Z")

</div>

I’ve been checked out from the drama since I deleted my reddit, but when I booted up the TR update and saw my previously-detected ftp drop by 30 watts I was tempted to pop some corn and log back on because I had a feeling it was going to wild over in the comments.

FWIW I’d been concurrent weight training and only doing maintenance cycling and I KNEW my old AI ftp was a gross overestimation, so I was pleasantly surprised when I saw it drop to exactly what my back-of-the-napkin math had me at.

---

<div class="post-metadata">

**Author:** ![vipepes](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/vipepes/32/85234_2.png) [@vipepes](https://www.trainerroad.com/forum/u/vipepes)\
**Post date:** [February 5, 2026, 11:03am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/44 "2026-02-05T11:03:04Z")

</div>

I have been training with around 310W FTP for some time now after completing a 20 minute test in Coll de Rates with 20min average power of 326W (also my all time best). My best 1 hour power is under 300W.

The first new AI FTP estimate for me is 352W. So at least for me, FTP and the new AI FTP are totally not the same thing.

---

<div class="post-metadata">

**Author:** ![El\_Jefe](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/el_jefe/32/7746_2.png) [@El\_Jefe](https://www.trainerroad.com/forum/u/El_Jefe)\
**Post date:** [February 5, 2026, 10:41pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/45 "2026-02-05T22:41:28Z")

</div>

> [@TxToast](#):
>
> For example, McGregor isn’t “3x20 minutes at 229 watts”, it’s “3x20 minutes at 88% FTP” and the wattage of those intervals is determined by… my FTP, that the system supposedly doesn’t know or care about?

Who says the system isn’t saying “this user needs to do 3x20 @ 229 watts to be productive today” and then hands that decision off to another routine to find a workout that equals those intervals @ your “FTP”. The AI didn’t care about “FTP” just watts and duration, which is what Nate is saying.

---

<div class="post-metadata">

**Author:** ![txrr9876](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/txrr9876/32/12908_2.png) [@txrr9876](https://www.trainerroad.com/forum/u/txrr9876)\
**Post date:** [February 6, 2026, 3:03pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/46 "2026-02-06T15:03:02Z")

</div>

I’m having a very similar experience

---

<div class="post-metadata">

**Author:** ![dolittle63](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/dolittle63/32/10249_2.png) [@dolittle63](https://www.trainerroad.com/forum/u/dolittle63)\
**Post date:** [February 7, 2026, 6:30am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/47 "2026-02-07T06:30:20Z")

</div>

Here is my suggestion to stop the fuss. at time of detection, create a button that automatically toggles the workout that day to minutes…. 15 min warmup, 60 min at the suggested FTP and 15 min cool down. Then there is nothing to debate about, its right or wrong if you really care that much ;-")

With that said, hey TR, please make a workout called “Hour of Power” with a warmup and cooldown preloaded as an option.

---

<div class="post-metadata">

**Author:** ![Twowkg](https://avatars.discourse-cdn.com/v4/letter/t/ce7236/32.png) [@Twowkg](https://www.trainerroad.com/forum/u/Twowkg)\
**Post date:** [February 7, 2026, 7:12am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/48 "2026-02-07T07:12:28Z")

</div>

Why would you do an exactly 60m effort to determine ftp?

---

<div class="post-metadata">

**Author:** ![AldridgePrior](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/aldridgeprior/32/139066_2.png) [@AldridgePrior](https://www.trainerroad.com/forum/u/AldridgePrior)\
**Post date:** [February 7, 2026, 10:32am UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/49 "2026-02-07T10:32:16Z")

</div>

> [@dolittle63](#):
>
> With that said, hey TR, please make a workout called “Hour of Power” with a warmup and cooldown preloaded as an option.

Do you have access to TR’s (legacy) “Teams” feature [NB. newer TR users don’t…]

If so, take a look at the “Team Gold” Team, which contains a bunch of “Kolie Moore”-style FTP testing workouts that someone created, based on this article’s content:

> **[The FTP Test: Physiology and New Protocols](https://www.trainingpeaks.com/blog/the-physiology-of-ftp-and-new-testing-protocols/)**
>
> The standard 20-minute FTP test might not work for all cyclists. Using WKO4 for deeper analysis, Kolie Moore explains new protocols.

---

<div class="post-metadata">

**Author:** ![Fats](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/fats/32/155745_2.png) [@Fats](https://www.trainerroad.com/forum/u/Fats)\
**Post date:** [February 7, 2026, 1:39pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/50 "2026-02-07T13:39:09Z")

</div>

Personally, my advice would be to ignore the noise. It’s your journey, embrace it, enjoy it. While others are moaning, you’ll be the one embracing the changes and getting the results.

---

<div class="post-metadata">

**Author:** ![Rizzi](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/rizzi/32/17674_2.png) [@Rizzi](https://www.trainerroad.com/forum/u/Rizzi)\
**Post date:** [February 7, 2026, 2:29pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/51 "2026-02-07T14:29:19Z")

</div>

> [@vipepes](#):
>
> I have been training with around 310W FTP for some time now after completing a 20 minute test in Coll de Rates with 20min average power of 326W (also my all time best). My best 1 hour power is under 300W.
> 
> The first new AI FTP estimate for me is 352W.

That’s wild. Keep us updated on your training, would be interesting to see your workouts. Also maybe share your all time and recent 8 weeks Power Duration Curve. Do you have strong short 5-10 minute power?

I had a theory here:

> [@AI FTP Prediction question - It felt disheartening to me](https://www.trainerroad.com/forum/t/ai-ftp-prediction-question-it-felt-disheartening-to-me/110876/32):
>
> Just to give you another perspective that was not mentioned here before: I also think that gaining that much is unlikely but we have to keep in mind that TR AI FTP it’s based around performing a threshold 3 workout… I think of 3×9 minutes intervals. So if you are high glycolytic athlete (for example, the kind of athlete that get an overestimated FTP based on 95% of 20 minute test and should better use 90 to 92%) and have your strength and power duration curve at 5 to 15 minutes … the gains he…

But at least your 20 minute power of 326 wouldn’t support that 352 value.

---

<div class="post-metadata">

**Author:** ![Trix8806](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/trix8806/32/155132_2.png) [@Trix8806](https://www.trainerroad.com/forum/u/Trix8806)\
**Post date:** [February 7, 2026, 2:37pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/52 "2026-02-07T14:37:30Z")

</div>

One could just do a “Free Ride” of a set length to determine their max power at whatever duration they choose.

My biggest qualm with the new FTP calculations is that since joining, I, like most, used FTP as one of the main metrics of my progress. I have a few durations which I have set 100% effort PR’s, but not as frequently as the TR updates. Fortunately last year I did an indoor ride where I did all out max efforts on a few shorter duration. I can find that ride, retest, and compare. I’ll probably start doing similar on some longer durations so I can more accurately measure progress.

---

<div class="post-metadata">

**Author:** ![jaredstrauss](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/jaredstrauss/32/157837_2.png) [@jaredstrauss](https://www.trainerroad.com/forum/u/jaredstrauss)\
**Post date:** [February 7, 2026, 7:00pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/53 "2026-02-07T19:00:35Z")

</div>

If you’re not using the FTP to apply for a pro team then who cares? It’s used to plan workouts on Trainer Road so whatever it thinks is best.

---

<div class="post-metadata">

**Author:** ![jkc](https://avatars.discourse-cdn.com/v4/letter/j/a4c791/32.png) [@jkc](https://www.trainerroad.com/forum/u/jkc)\
**Post date:** [February 7, 2026, 7:11pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/54 "2026-02-07T19:11:17Z")

</div>

TR is just using the inflection point on your power duration curve to define your FTP. It’s the second of “The Seven Deadly Sins” by Dr. Coggan, posted to the wattage forum in June 2004: estimating FTP from the power distribution profile from multiple rides. It’s #2 because it’s second in terms of certainty, whereas #7, the best predictor of performance, is performance itself. TR didn’t do anything new, but created an algorithm to find that reflection point instead of relying on experience. To have a good estimate, you need to feed the model. Feeding the model means you need to have representative data across your PDC, points before and beyond your FTP, else it’s garbage in, garbage out. I wonder how close the AI FTP is to the 95% of 20-minute MMP today? I haven’t used it recently, but my impression from the initial rollout, it was 95% the 20-minute MMP. It is a faulty assumption because it ignores the 5-minute blowout that the test creator deemed critical in determining FTP when using an abridged version of the test.

---

<div class="post-metadata">

**Author:** ![dolittle63](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/dolittle63/32/10249_2.png) [@dolittle63](https://www.trainerroad.com/forum/u/dolittle63)\
**Post date:** [February 7, 2026, 7:14pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/55 "2026-02-07T19:14:41Z")

</div>

Functional Threshold Power (FTP) is **the highest average power, measured in watts, that a cyclist can sustain for approximately one hour, representing their maximal lactate steady state**

Everyone bitches about the “validity” of their FTP when in reality, i bet none of them ever measure the true number and always to Ramps, 2x8s, 20 mins, etc. Sort of not really worth the discussion if the number works for the training protocol.

---

<div class="post-metadata">

**Author:** ![dolittle63](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/dolittle63/32/10249_2.png) [@dolittle63](https://www.trainerroad.com/forum/u/dolittle63)\
**Post date:** [February 7, 2026, 7:15pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/56 "2026-02-07T19:15:40Z")

</div>

I have had TR since 2020, So i assume I am “legacy”. I will read this for sure, thanks!

---

<div class="post-metadata">

**Author:** ![patrickhill](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/patrickhill/32/12893_2.png) [@patrickhill](https://www.trainerroad.com/forum/u/patrickhill)\
**Post date:** [February 7, 2026, 7:20pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/57 "2026-02-07T19:20:04Z")

</div>

It’s a bit of a bummer that there was a discontinuity of AI FTP detection with this launch. Mine was exactly 5% different, so I can adjust. But whatever your delta was and regardless of whether TRs number matches with traditional definitions of FTP, I would think you could use the new TR AI FTP detection (not prediction) as a solid way to track fitness going forward?

---

<div class="post-metadata">

**Author:** ![kevistraining](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/kevistraining/32/171237_2.png) [@kevistraining](https://www.trainerroad.com/forum/u/kevistraining)\
**Post date:** [February 7, 2026, 7:37pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/58 "2026-02-07T19:37:19Z")

</div>

> [@jkc](#):
>
> TR is just using the inflection point on your power duration curve to define your FTP

I don’t think this statement is true at all. Doesn’t seem to rely on any kind of max efforts.

Instead the new AIFTP seems to be determined by what workouts an athlete can do.

Process used to be:

Assess fitness (test/AI) \> define FTP \> build workout based on FTP

New process is:

Assess fitness (AI) \> define workouts \> display FTP

---

<div class="post-metadata">

**Author:** ![Torgo](https://sea2.discourse-cdn.com/trainerroad/forum/user_avatar/www.trainerroad.com/torgo/32/7721_2.png) [@Torgo](https://www.trainerroad.com/forum/u/Torgo)\
**Post date:** [February 7, 2026, 7:54pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/59 "2026-02-07T19:54:07Z")

</div>

Yes, it was just a label that was applied to a number they got after looking at 40K TT power data. Back in the day these were pretty common so was a reasonable number to base training on.

I would be good money a majority of people can’t ride a 40K TT outside at their current FTP (me included) 🙂

---

<div class="post-metadata">

**Author:** ![Twowkg](https://avatars.discourse-cdn.com/v4/letter/t/ce7236/32.png) [@Twowkg](https://www.trainerroad.com/forum/u/Twowkg)\
**Post date:** [February 7, 2026, 8:00pm UTC](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694/60 "2026-02-07T20:00:00Z")

</div>

\*Approximately \* 1hr, not 1hr exactly which was my point. It’s a proxy for MLSS / maximal metabolic steady state.

I bitch about it because people get it wrong all the time lol. It’s really much simpler to just do a long form test and actually know what your ftp is.

[Previous page](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694.md?page=2)

[Next page](https://www.trainerroad.com/forum/t/trainer-road-ftp-vs-ftp/110694.md?page=4)
