This could be because I spend all day looking at code and numbers and trying to understand the link between, but I do feel that if it was completely auto (like intervals.icu eFtp is entirely auto) some sort of feedback along the way would be nice, intervals has a “your eftp from this ride was x, staying at y” on every ride, not saying that it has to be every ride, but some sort of message at the end of the week, along the lines of “ftp on track, within progression level tolerances, keeping at x” I feel would be helpful rather than just waiting 28 days for magic to happen (and nothing happening)
5 Likes
I fully second the idea of some form of messaging in the event of “No FTP Change” so we get some feedback. The absence of info leads to questions of whether or not something is working vs broken. A flag that says something would be use for those of us with minds that wander 
28 Likes
I know you quoted me, but I am not the right person to answer those questions.
It’s hard for me to make an informed decision without trying “auto” first. That would be my preferred option but if I felt like it wasn’t for me it would be nice to default back to manual.
I would also like to be able see estimated FTP changes more often than every 4 weeks. Ideally I’d like to see this like other platforms provide modeled FTP - you can look at it whenever you want.
In particular, if/when I do any kind of FTP test, I very much want to know what the TR estimated FTP is at that point in time, as this is crucial to helping me know how it is working for me as an individual.
This is what really matters to me - ie how it works for me, not how well it works on a population level.
I also think 2 different things are being conflated in this discussion:
- My actual physiological FTP, aka ‘real’ FTP. This is likely to change relatively slowly, and is unlikely to need to be updated more often than every 4 weeks. More frequent updates could be useful, maybe in situations where a very fit athlete is coming back from longer break, or a new rider is seeing fast initial gains.
- My modeled FTP based on recent ride data.This can change rather quickly, as ride type changes, motivation changes, etc. I could easily see fairly quick changes in the estimated FTP as the type of rides done changes. I think a 4 week minimum interval between estimate updates is likely to delay useful updates when there are significant changes in ride/workout types for a rider.
Also, what is the point of the auto mode ‘checking after every ride’, if you won’t get updates closer than 4 weeks apart? Am I mis-understanding how the 4 week limit applies to auto mode?
2 Likes
I think you’re misunderstanding, or I might be too. But it sounded like AUTO would update more frequently than four weeks if the change reached whatever the markers are for a bump.
2 Likes
right here:
1 Like
I’m sticking with fitness tests but I’m happy to see TR making progress with AI FTP Detection. My only request would be a way for AI FTP Detection to work better with outlier athletes. I know we’re a very small minority but I’m keeping hope alive!
But you’d name is Chad and all Chads are the same with knowledge, so…
When FTP goes up PL levels go down causing TiZ to decrease for TR workouts as they stand now. Workouts are done at a percent of FTP and the specific workout picked is based on your PL level. Two people with very different FTPs but the same PL level are likely to do the exact same workout. (the power levels in the workout are a percent of FTP so they would be doing them at different target powers)
1 Like
Taking a stab at this, ignore “other people” and just focus on a single rider.
-
Presume a rider has trained for several weeks at a given FTP. They get an FTP update of 4% higher from TR, and then need to continue training.
-
Presume no PL change/drop, which would mean the rider demand is higher from the elevated FTP even if the TIZ is the same (from workouts at the same PL).
- Maybe this would work for that rider, especially if they were rating recent workouts in the 1-3 range.
- But more often than not, I would expect people to struggle a bit with workouts at the higher FTP at the same PL.
-
TR drops PL’s because it’s not necessarily likely that the rider can take the new FTP and perform the same workouts at the same PL’s. This is a practical solution to the reality & anxiety we have all seen or had after getting a higher FTP within a TR program. I and many others often struggled with the workouts following an FTP bump within the old TR plan setup.
-
AT, and the PL’s that are part of the whole picture, deliberately step back a bit to make the transition to the new FTP a more gradual one that is easier to handle than the ways of old.
As we’ve seen, TR was tweaking the amount of reset in the PLs since they were initially released. We saw massive drops in the early times, but I am seeing much more gradual and practical drops in PL’s. These make a more appropriate reduction in effort for near term workouts and is far better than the constant “upwards and onwards” that was present before AT.
7 Likes
2 Likes
The TR MacOS app updated for me this morning, and it provided the instructions on annotations coloring I was missing: the calendar now remembers the color you used for a specific annotation (e.g., Notes, Time Off, etc.) and defaults to that color the next time you use that specific annotation.
4 Likes
Just for additional clarification. Either way, we could drop a Ramp Test on the calendar and use AI FTP as long as there is enough time between the last update of FTP.
Does not really matter for me as I put in another post yesterday. Pretty sure I am getting close to my max FTP. That is unless I am suddenly willing to hurt a lot more, either in ramp tests or training.
Much more appealing than what I understood it to be.
You could have a feature where if people click on manual FTP, the software could say something like, “look you could reasonably set your ftp to x but let’s be honest, you are at Level 3 in most progression levels. It is clear that you will get better progress by working through the levels. And your next update is only 17 days away…” Or something like that (or perhaps more snarky) and most of us will just get it and keep progressing.
1 Like
Intervals.icu does this with eFTP updates after relevant workouts, although presumably TR has a larger dataset to work with to get more accurate FTP estimates. I find intervals.icu to be lower than my ramp test results or TR’s auto detection, but I don’t get too excited about the gap. Absolute accuracy is not relevant. Having a relative benchmark is.
2 Likes
How will this handle FTP decreases when somebody takes time off or greatly reduces volume? Will the adjustment happen in a similar fashion to the increases?
Asking for a friend 
2 Likes
How would it work after an off-season or a long break ?
I’d like to think AUTO could help a lot there. Still Manual is a way to check (and correct sometimes) the automatic value if I think it doesn’t work well. Why not keep MANUAL every 14days?
Also I’d like as a user to decide whether to prioritize FTP up or PL progression out. That could be a user setting.Yes it depends on how far away I am from my A race but there are other factors too, how tired I am , what kind of stressor at work, …
This would be a brilliant update Nate. Im going to do some more TT’s and would find accurate FTP so useful.
It is a functional test not a physiological test.
1 Like