I was very excited to go back to an older activity and see this metric, but on my ride yesterday on the same segment I did not see it again. Is HRRc gone?
I’ve read some of the previous posts on this, and the articles linked, and personally I would have looked at a different implementation.
Coming from engineering background, whenever I see an exponential decay I think in terms of time constant and response to a step function, not really as drop over time, because start/end points and steady state values differ.
So if you try to fit HR data following any clean interval to and exp(-t/tau) and best fit for 'tau" you would get a time constant that tells you the rate of recovery.
BTW same approach could be used to measure the ramp rate at the beginning of an interval.
Also you wouldn’t be limited to a 60s recovery time, as this could be calculated also for shorter recoveries.
It’s not gone You have to reach at least threshold HR to get an HRRc measurement and you didn’t quite get there on that ride. It keeps the best (highest recovery) from all the qualifying drops. I implemented this after reading some stuff about LSCT tests and was hoping to get something that could be used in a similar way. Thats why I chose 60s and an easy to understand “how much did HR drop” result.
So are you suggesting that I should calculate the highest rate of recovery for the whole ride?
I was at/above LTHR for the duration of the climb, 46+min, but really close, I guess I need to be above it for a little while. FTP bumped up 15W as a result of this effort, definitely significant for me.
Are the HR zones updated dynamically or fixed based on my input in settings? (I realized I haven’t updated those in a while…)
I’m not suggesting anything really, but speculating that would be interesting to look at that data (and I haven’t seen any other tool in WKO/GC to do that.)
One more thing to think about (which always makes me question W’ models) is how the recovery time constant of W’ is set… I really like how W’ and HR recovery at the end of the steady effort almost perfectly mirror each other.
If Intervals.icu thinks your LTHR has increased it will flag the activity and you can click to update your HR zones:
So fixed but you get an alert for LTHR changes. If you exceed your set max HR the HR data will be fixed assuming its a HR strap anomaly. You can choose to bump up your max HR instead.
I must try make some time to play around with modelling HR recovery like you suggest.
Tx. You can start with 90% of your max HR for LTHR and then see how it goes. Intervals.icu will give you a notification if it thinks your LTHR has gone up … it uses 98% of your average HR for 20m as a LTHR measure.
You just have to reach threshold HR to get an HRRc measurement. 1 sec is enough.
You have to reach at least threshold HR to get an HRRc measurement and you didn’t quite get there on that ride. It keeps the best (highest recovery) from all the qualifying drops. I implemented this after reading some stuff about LSCT tests and was hoping to get something that could be used in a similar way. Thats why I chose 60s and an easy to understand “how much did HR drop” result.