AI FTP Detection Update

Regarding the name change, though, I have an idea. If you want to improve upon FTP, then you should name it what it is: estimated lactate threshold. This disambiguates it from wrong interpretations of what FTP supposedly means (e. g. 60-minute max power), it is clearer and most importantly, it is scientifically accurate.

And I assume it is internal for good reason: it is not clear for whom this metric works well and for whom it doesn’t.

The difficult thing is that I can easily see several layers of metrics, e. g. metrics to compare seasons or so. E. g. it’d be nice to have an end-of-season report where e. g. I’d see stuff like total hours, total hours in zones, average hours per week, etc.

For example, the metric I’d probably want to know in addition to FTP is endurance/repeatability. I know that this isn’t very easy to quantify or score, but endurance is essential and typically forgotten by beginners who tend to focus only on FTP. But you could first create an absolute weighted score that measures how many intervals and how many minutes you manage e. g. at threshold or VO2max. Then you rank all athletes. You don’t actually expose this absolute score to the athlete, though, but you compare athletes relative to one another. E. g. you could compare athletes of the same gender with either the same FTP or W/kg with one another; perhaps you throw in age as well.

Once you have these metrics, you could classify and rank athletes. Not that you’d want to expose that I am in the bottom 20,000s of sprints or so, but you could categorize people and build a profile. You could identify relative strengths and weaknesses, and then suggest e. g. two training plan alternatives, one to improve your strengths, the other to address your weaknesses. I am sure you have thought of this as something you’d like to have in 10 years or so.

Thanks for sharing, that’s pretty neat. Did you feel the score was accurate and informative? (I remember that you had a training hiatus before Cape Epic and didn’t feel you were anywhere near the top of your game. Just subjectively from how you spoke of your fitness, I would have expected a larger drop in fitness score.)

Yeah, that’s the big one for sure and really needed to get an accurate picture of the fitness of an athlete. Especially when it comes to endurance rides, TR has me way too low, because I don’t do long trainer rides in the summer (I don’t have more than 2 hours in the morning, 2 hours is stretching it, and on the weekend, I wanna be outdoors).


I know you said in previous that you don’t use TR workouts currently, but if you were to feed the Machine information through the surveys it would customize a plan around this. So your PL level would be higher for anaerobic workouts while your threshold workouts would be lower relative to PL.

It takes sometime to get acquainted but will build itself to best fit you.

1 Like

@Nate_Pearson I would find an AI generated hour power approximation extremely useful. I think that is a great idea.

1 Like

Why don’t you call it TTP (Training Threshold Power)? That way it. Is divorced from FTP entirely


Makes sense to me then. Keep it relatively aligned but still tag it different enough that it has unique identity to serve the purpose. I’m all for spending that dough on real stuff vs marketing and education that can be minimized or avoided entirely. :+1:


Yah. And we’ll backfill PLs all the way back too so you can see it over time. Bit we are prioritizing WLV2, AI FTP, Red/Green all before that.


You’re like a former bodybuilder or power lifter, right? Your IG picture makes it look like that. :muscle:

When were you doing this? Right now it looks like threshold is way above anaerobic.

You’re hitting on a lot of stuff!

PLs do give you a measure of repeatability and endurance. That was one of the goals in creating them.

And you’re also very smart about the ranking system. I would like to know how my fitness score (or another score) ranks against all athletes, or athletes in my AG or area or something. Even if I was at 4000 and then went to 3995 I’d be happy :-D. Those 5 people can eat it! :laughing:

And Peloton has shown us this is very motivating.

For the FTP stuff. It’s not about being as accurate as possible with the name so that the top posters (and a few podcasters) are happy. We need to develop stuff for millions of athletes without diluting the training effectiveness.

If this was 2011 or Zwift didn’t have so many users we could probably rename it. But “FTP” is so ingrained in so many people who don’t care about the specific definition but want a number to keep improving.

I’d love to give them more motivation, and finer control with PLs and eventually the fitness score. Then to rank those against other athletes on TR.

So naming it has to consider marketing and the cost of educating millions of people about a new term. I think it makes the most sense to have a support article for the die-hards, then start to introduce PLs and Fitness Score to more athletes.


Oh, and yes, the fitness score was a good representation of my fitness.


It wouldn’t even be AI-generated, just math based.

If we used AI we’d have to train it against people’s ACTUAL hour PRs. Which very few people do…and everyone would have super low estimated hour powers.


That’s a huge marketing effort that would cost millions for it to take hold in cycling.


This was about 3 or 4 months ago, on my second block of threshold work now and moving on to VO2max in a couple weeks.

I think “Training Threshold Power” (TTP) conveys the same change but makes more of a distinction from FTP which is needed for what you want to accomplish. Or you could think of it as “Trainerroad Threshold Power” if you want, still TTP. including the “F” means people will just say FTP and think of it that way


I see I’m not the only one to suggest this.

This is such a good comment, I hadn’t thought of that. Zwift does the same, the gamification is meant to make people more consistent. This is a double-edged sword, but can ultimately be used for good.

True. Which is why I wrote to stick with it, the suggestion was just in case you wanted to rename FTP.

Here is a metric you can implement very easily and which will have an impact: workout streaks/consistency. Count the workouts a user has successfully completed in a row and give a percentage of successfully finished workouts in a training plan so far.

Workout streak: 15 workouts in 3 weeks
Consistency so far: 95 %

You could also give more positive feedback after each week by displaying a message like “Finished another week as planned! Keep on going! :+1:

The more I think about it, the more I am convinced that this is something you should implement and implement ASAP:

  • It is easy to implement. In fact, I assume you are already tracking these metrics — along with many others — internally
  • The metric is easy to understand. No ML is necessary.
  • The metric incentivizes behavior that we know is one of the most important predictors of getting faster: consistency.
  • The only difficulty I see is how you handle failure — you want positive reinforcement and not discourage users.

The bit that i found extremely interesting is the idea about balancing TIZ vs ftp wattage number. As someone who believes the best way to increase ftp is to develop a large TTE, I would plan to try the 40k tt specialty and sustained power build…

That’s the point: it doesn’t need to take hold within cycling generally, just be clear to TR users what the number that the TR AIML system is determining can and cannot be used for. It’s completely analogous to Workout Levels / PLs which have no meaning outside TR. These are all just metrics that the TR system uses to give the user a series of workouts to help the user meet their goals.

I agree with those who state that you can stick whatever you want in front of FTP, and all people will see is the FTP part.


I don’t know if that is a good idea, because you are overlaying a mixing a training goal with your particular training methodology that you believe is best suited to achieve that goal.

The way AT works right now is that it generates its own training methodology from the goals implicitly contained in the various training plans. So if you opt for a 40k TT training plan, then AT might not choose your training methodology. I am not saying you are wrong and TR is right, I’m just trying to disentangle goal from method, and concluding that you should not assume AT will follow your method.

1 Like

Point well taken… generally i try to just apply principles i learned from my time when i was coached… it may not be quite as good as continuing to work with a coach… but just being able to effectively balance my fatigue is a big win… does AT offer this?

I think that is totally fair, some people respond better to some methodologies than to others. And the placebo effect is real — the body is more likely to react the way you want if you believe it will. (Just to be clear, this is not meant as a putdown or so, the placebo effect is measurable.)

At present probably not in the way you want. At some point in the future I’d like to be able to e. g. have some influence on how I want to improve my power profile and how many days I want to train for and for how long. This way you could have AT create a plan for you with more endurance rides in the mix (i. e. less intensity).

Some of this is in development (@Nate_Pearson’s ship is leaking from the top), but likely won’t be available for some time. They first have to get scoring of outdoor rides out and then might have to iterate on that.

1 Like

Thanks for answering these questions again, Nate. I know they have been asked and answered previously. I was just trying to express that these are the issues we (or at least this user) care about, more than the other concerns you mentioned. I should have expressed it differently.

1 Like