I know this is a strange question, but does Adaptive Training use heartrate information in determining adaptations and progressions?
I’m asking because I record my TR workouts on my Garmin Fenix as well as in TR. Many times, I just use the wrist heartrate provided by the Fenix instead of putting on the HR strap. In those situations, unless I ask the Fenix to broadcast heartrate, the heartrate info never makes it into TR.
For my own purposes, I use Garmin Connect as my central hub of information. So missing info in TR doesn’t bother me. That said, I don’t want to lose the benefits of adaptive training, progression levels, or TrainNow.
Correct, AFAIK. TR is not using HR at this time for workout related to Adaptive Training.
HR is currently used for TSS estimation (if you have that setting turned on) for any workout (inside or outside) that doesn’t have power data.
Based on comments from Nate in a couple of recent casts, I expect that HR will be growing in their use, but we don’t really know specifics at this time. Nate’s basic summary was to add HR now so it can be leveraged in the future. That implies there may be some backward use as well as whatever future use.
I’m kind of surprised HR hasn’t already been used in at least ftp estimation, tracking efficiency factor changes over time would certainly help inform any estimates, I’d think.
WindWarrior Adaptive Training has used HR since 2016
On a serious note, unlike power zones from recovery to threshold, HR zones do not have the same consistency between athletes. And TR has long promoted power based training, starting with virtual power.
I’d have to go back and recheck, but I do think TR may have added this or it’s actually part of the new FTP Detection. It was part of the back and forth between Nate & Amber on one cast, and I think they recovered it in another one, but I don’t remember the specifics.
As I said, it sounds like Nate is more on board with HR now so I expect it’s use to widen over time.
I’ve been running a rolling 30 day average of EF and on 2/1 it was 1.55 and today it’s 1.57. That’s gotta be good for something. But generally when I do sweet spot I get a good feel for progress based on how my Hr trends around 90%
From time to time I’ll glance at EF, however I come from the Friel school of thought and associate EF primary use with steady-state aerobic efforts (zone2 workouts). Comparing EF of all workouts doesn’t make sense in my experience - tried multiple times, I don’t find any value from that. I’ve started tagging my zone2 workouts in WKO to make it easy to only display EF trends for z2 workouts.
For HR trends on intervals I think it helps to do the same interval work week after week. I know a lot of people like to mix things up because of boredom, but that just makes it difficult to compare power-to-HR on say sweet spot intervals over time.
Seems like HR recovery during between interval rests would be really valuable for the AI to push towards longer intervals and shorter rests at the right rate over the course of a plan.
I don’t find EF that valuable when comparing one ride to another, unless duration/VI/IF are all pretty similar, though do compare it from one interval to the next within a ride for a sense of how much decoupling I had and where it hit. I’m guessing that data would also factor into improved plan adaptation.
In any event I’ve been really impressed with the work so far and am excited to see what the future holds.
Good to see this is on the roadmap as I’ve been curious about this since I started and heard it was not part of AT. Is this just a product limitation/lower priority or are there scientific reasons for leaving it out?
I understand that HR varies from athlete to athlete and even day to day for the same individual but I imagine some trends emerge over time. Once they do, I always thought HR would be a more concrete metric than simply asking how hard the workout was. I’m more of a black/white personality so choosing something so subjective when AT is so smart always gave me a bad feeling.
This clearly isn’t my area of expertise though so I’m hoping someone can either confirm my thinking or tell me why I’m off…
I’m glad I’m not the only one to ask that. I was thinking 30 days of Language Lesson was a lot or maybe @hubcyclist is on the team (EF = Education First )
In general I don’t know have a good answer. I started looking at per ride EF in TrainingPeaks back in 2016. Maybe the paid subscription has it, or maybe Golden Cheetah or Intervals.icu or some other analytics package.
I’m using the WKO5 analytics package because I can easily do things like tag rides as zone2 and then modify standard charts to only show EF trends for zone2 rides. In my data for this season (started Sept 1), there is a (small) difference between EF trend line for all rides vs zone2 rides.