Athlete Login
5% or 10% drift for AeT test and adjusting test duration
- This topic has 6 replies, 4 voices, and was last updated 1 year, 9 months ago by
.
-
Posts
-
After a botched attempt at a treadmill heart rate drift test this morning, I was looking into the possibility of performing a curve fit to salvage some partial data. I revisited the various articles on performing an indoor heart rate drift test and I think there is an inconsistency between the acceptable range for heart rate drift per hour and the way the data is analyzed. In the most recent article on heart rate drift test it says that if your heart rate drift is less than 5% per hour at a steady pace then you are below your AeT. However, when analyzing the data it has you take an average of the HR for the first half hour of the test and compare it to the average HR to the second half hour of the test(small aside, it would be nice to have examples of tests that were correctly performed at Aet as well as above Aet for reference in the article). When comparing these two numbers we are looking at the percent drift per half hour not per hour. If we wanted to know heart rate drift per hour we would neet to look at the start value and the end value which seems like it should be closer to 10%. I realized this may be a bit nitpicky as no one seems to have had an issue with this before, but I have seen some recent discussion about adjusting the duration of a heart rate drift test and I think it is important for understanding how the numbers work in order to make that adjustment. What is the recommended method for adjusting for a different duration heart rate drift test? Is it safe to assume that heart rate drift is fairly linear for tests at or under 1 hour long? So for shorter duration tests we just compare the first half average HR to the second half average HR, and then divide that percent drift by the test duration in hours? So if the test was a half hour and we get a 2.5% drift comparing the two halves, then the hour adjusted drift would be 2.5%/0.5 =5%?
Another question that have come up as I have been thinking things over: Is there a point at which heart rate drift leaves the linear region and begins to flatten off? I would think that your heart rate couldn’t drift 6-10% per hour for multiple hours on end. Has that been looked at before?
I think you are referring to this article –
https://evokeendurance.com/our-latest-thinking-on-aerobic-assessment-for-the-mountain-athlete/
Here is the training peaks article on Pa:Hr (Aerobic Decoupling used for outdoor tests) that also explains the calculation
https://www.trainingpeaks.com/blog/an-introduction-to-more-trainingpeaks-metrics-part-two/
For the heart rate drift (indoor test) calculation, instead of just taking two data points (beginning and ending HR), we are using the average heart rate of first half and second half as mentioned in both the articles above.
I would recommend doing the test again after a good warmup. If you want, you can make your workout public and share the link here.
the key is warmup. for me, that means at least a 15 min warmup including 5 mins of being at my target HR for the test. this is my last drift test done a couple of weeks ago on a treadmill. My preference is a moderate pace done at 2% incline. the graph shows a 15 min warmup and then a 58 min drift test. my target AeT was an avg of 145, my actual avg was 146. I don’t worry if my HR is bouncing around between 145 and 150…which happens naturally based on the song I’m listening to, what I’m thinking about, did I take a sip of water, etc. if my HR gets a little above my target, I focus on relaxing and it drops a few beats. I never adjust my speed or incline during the test. I suspect that my test would have been just as successful at a target of 150, so that will likely be my target for the next test
Hi pvoosen,
I thought about your post and I would agree:
When performing a 1 hour drift test and comparing the mean HR of both halfs, then the ratio it is equivalent to the HR drift per 30 minutes.The duration of the entire test and specifying the slope of the HR drift curve seems to be a source of confusion. I hope I can illustrate this with the above chart. In fact, we are always talking about the same slope but in different units.
As Shashi said it is more reliable to compare two means than to compare two single data points. At TrainingPeaks they say “Aerobic Decoupling values for efforts under 20 minutes in duration aren’t as valid” (https://help.trainingpeaks.com/hc/en-us/articles/204071724-Aerobic-Decoupling-Pw-Hr-and-Pa-HR-and-Efficiency-Factor-EF-). So in my drift tests I compare means and try to go as long as possible. Afterwards I adjust the percentage according to the duration of my test, to be able to compare drift tests of different duration. I assume TrainingPeaks does this duration adjustment of their pa:hr metric under the hood.
I hope I could help a bit. 🙂
Christian@Christian, Thanks for providing the useful graphic although I think it may be more accurate to say 5% drift per 30 min instead of 5 bpm. This graphic shows exactly what I was talking about. I have spent the morning reading through several training peaks articles and I am not sure it is safe to assume that training peaks adjusts pa:hr based on test duration.
Aerobic Decoupling compares the Efficiency factor from the first half of an activity to the second half. A smaller change in EF (less than 5%) from the first half to the second half may indicate improving aerobic endurance.
Based on the description, it seems like the first half of the workout is compared to the second half and a direct percentage is calculated without adjustment. I think this implys that for an hour long aerobic assesment showing 5% pa:hr over the entire duration, if you look at the first half of the workout you should see a value of 2.5% assuming that pa:hr is fairly linear.
Based on what I have read in training peaks article, it seems that a value of less than 5% for Pa:HR for any duration is the magic number for an effort not being “too hard”. So should we look for a 5% Pa:HR regardless of test duration or do we need to manually adjust the percentage to fit the test duration? Logically I would think that some compensation needs to be performed for the test duration, but am just looking for confirmation.
Hi @pvoosen, thanks for your reply! It helps me too to think things through and understand these concepts precisely.
- I decided to write 5 bpm instead of 5% because 5 bpm of 95 bpm is 5.26% and 5 bpm of 100 is 5.00% … so the percentage value changes as you move the slope triangle along the line although the slope is the same. I wanted to refer to the bare slope without messing about with percentages.
- You may be right and TrainingPeaks does not adjust for duration. It was just a guess and after seing your research on the topic I would take my assumption back.
- But I completely agree with you, some compenstation for duration should be performed! An activity that produces 5% fatigue in one minute is definitely of different intensity than an activity that produces 5% fatigue over two hours. So to my mind only fatigue per duration can serve as a reasonable indicator for threshold intensity.
- You must be logged in to reply to this topic.