Since the data will still be in TR the AI will take that and adjust your training for you. So no need to do anything.
4 Likes
Just got served up Katahdin -2 (VO2 float sets) by the AI and it felt perfect
4 Likes
Yay! That’s what we want!
Probably very different from person to person. About different indicators of form, capability.. it might be a lot more interesting to not so good cyclists, as I am too😀, then all the better ones. Also difference is easier to achieve and so provide more motivation.
I would be absolutely unsurprised by the amount of Claude you or any software organization utilizes. It’s a huge force multiplier for senior developers. Mostly I worry that this will lead to not enough junior devs being hired and then there not being any senior devs in the future as a result.
5 Likes
Would love this feature. With the data you have on athletes, do you think you could provide a score for an athlete’s potential as well? I can see this being a little more difficult though
1 Like
If you do this, have a graph that is age-adjusted. WKO has something similar but not identical.
4 Likes
How comparable is the “new” FTP number and the AI FTP Detection of old? For example I know that I’d be able to make the front group of a specific race with an FTP of 300 based on previous years, but now with this updated FTP model will that 300 still stand?
Edit: It seems like it’s “pretty close” based on Nate’s answer in a different thread.
1 Like
Fine! I’ll start using a heart rate monitor again! 
1 Like
You’ve probably seen me post this before, but I would love the have TR’s historical assessment of my FTP, and this new aggregate score, when it comes, to see how I’ve progressed over time.
6 Likes
Assuming you mean like pick a random tuesday and see what it thinks you could have done for FTP? They do seem to have that on the back end as Jonathon mentions he runs aiftp on the podcast all the time when analyzing someone’s questions. So adding a button to let users do it might not be a crazy ask. Someone will figure out how it would be problematic though.
1 Like
He did say pretty close in most cases, but the example he used me for moved it up by 21W (~6%). Maybe that was just for illustrative purposes or maybe it can vary that much. Either way, it seems to be more about the power than the back-calculated FTP.
2 Likes
I don’t know about potential; you’re all shining stars in my eyes. 
I think you could do a longer prediction frame with some Monte Carlo simulations to get an idea of how far you could go over a great time period.
4 Likes
How about a rank inside your age group?
6 Likes
We’ve always had this issue where FTP + PL is your fitness, so it doesn’t really change with this.
You might be at 300 FTP one time and do level 2 threshold, then 300 FTP and doing level 8 threshold. That’s a big difference.
I think a historic athlete score would fix that though.
3 Likes
I’m now smashing the refresh button on the Early Access Page. Let’s go! 
6 Likes
Nate, that would definitely work. If you just compare to the Pro (1000), then the graph is basically meaningless. If you have an additional graph that is age-adjusted or ranked with one’s age group, that can be motivating and/or satisfying.
4 Likes
Would you use something like compound score? Obviously maybe expand it a bit but similar concept?
1 Like
Would love that if it were elevation adjusted.
1 Like
Yah, that’s on the long list of things to do.
4 Likes