Dfa alpha 1 in Xert ebc

General
#1

@ManofSteele, nice article in pezcycling (How To Use Heart Rate Variability DURING Training – PezCycling News) on using dfa 1 for zone 1 training.

You managed to convince me to cough up the cash and try the hrvlogger app. It turned out my zone 1 was close (-10w) to my LTP, which i was impressed with despite testing on an outdoor ride. Fortunately I ride the same outdoor route 2x every day, so it’s going to be interesting to see how consistent that is over the course of a week, and over time with fitness.

So im now left wondering, is it wishful to think your article is a sign that a DFA alpha 1 field will be coming to xert ebc in the near future?

I sure hope so, as it’s a pain switching between the two apps (xert + hrvlog), and I don’t want to have to choose which one gets sole rights to my bluetooth hrm on zone 1 rides!

1 Like
Dfa a1
#2

If you have the Polar H10 then it has dual Bluetooth streams. You have to turn the 2nd one in in the Polar app.

2 Likes
#3

Hi @Jarradk ,

Thanks for reading my article! I hope you found it interesting and useful as you start experimenting with DFAa1!

At this time, no. The articles I write for Pez are typically related to (relatively) newer research studies that I find interesting. I think DFAa1 is one of the hotter topics in endurance sports right now, so figured it would be a good topic for the article. There are no plans at the moment to include any form of DFAa1 monitoring directly in Xert EBC or Xert Online. Seems most folks are using the HRVLogger or Runalyze to do some of the analysis with DFAa1.

As @carytb mentioned, there is a setting in the Polar app to duplicate the H10’s Bluetooth signal so you can pair to two apps simultaneously.

2 Likes
#4

Just wondering if anyone has successfully used the HRVlogger app with a Garmin Dual HRM? I got the app yesterday, did a test last night and again this morning using the protocol in the article and alpha -1 drops below .5 by the end of the 10 minute warm up!

I suspect something must be wrong as the wattage is waaayyy below my all day pace , which can mean up to 10 hours plus on occasions. It’s roughly about 70 watts below LTP and drops to around .3 about 100 watts below TP, so that can’t realistically be correct.

#5

Does seem unusual, maybe check the artifacts removed chart for evidence of anomalies?

I am on garmin dual and my 0.75 alpha is around 10-20w below LTP, but it fluctuates a bit from day to day.

I can get strange results if it’s within the first 0-15m of my ride where I haven’t got sweaty enough for a good/stable hrm reading, so I generally turn off the app and restart it later into the ride.

1 Like
#6

Very likely there are errors in the raw HRM data. Be sure you have the R-R interval correction set to ‘Workout’

1 Like
#7

Yes, I made sure that was selected along with 2 minutes for the duration. I’m giving the strap a wash now and will change the battery to see if that makes a difference. Something is off as my TP is 299, LTP 259 in Xert yet I’m below. 75 by less than 180w.

#8

Jarrod, thank you so much for posting this topic. The links within the PEZ article are themselves well done in my view, e.g., the commentary having to do with apparent increasing validity of the STRYD device, as well as those others that describe the very many sensors that can also be used. I had no idea it was possible to assess muscle oxygenation w/o invasive measurement.

FWIW, I got HRV Logger and I’m using it with my dual-band Garmin. BAM. Flawless data acquisition. I’m now trying to figure out how to mount my iPhone and my iPad so that i have concurrent displays of metrics available from the Logger as well as EBC.

Again, thank you very much, Anthony

#9

I also get results all over the place despite controlling for all recommended aspects… HRV is sensitive to many things (time of day, fatigue, caffeine) and people’s baselines are also very different. Some of those things may be dealt with by the detrended fluctuation analysis (don’t know the math) but suspect not everything is given variability if results. The whole thing is still really developmental, so would take with a grain of salt. The papers I’ve seen so far are really anecdotal I.e. one to two people, and I’ve seen coaches commenting on other forums that they wouldn’t rely on it at this stage due to mixed results.

Personally, for LT1, I would just stick to something like 90% LTP or the talk test to go old school

3 Likes
#10

I’ve found that my 0.75 alpha HR and power (recorded via H10) varies from day to day depending on how strong/fatigued I feel. I think this is right as I’ve read and heard elsewhere that even TP is not a figure but a range depending on the day and what you did the day before. I guess just like HRV is also a range and I might look at how my 0.75 alpha follows the daily HRV score. I think the trick is to stay above 0.75 even when you are really fatigued, but that may mean pedalling with less than power than normal.

#11

I’m on the side of waiting for more detail on it before changing things. There isn’t any research to suggest following DFAa1 regularly rather than just following consistent power targets results in more improvement. One could argue that the DFAa1 values could be suppressed by other stressors not related to cycling and that you could sustain your regular power targets. (Think about showing suppressed values from having done a hard upper-body workout one day but being able to do lower-body the next day, for example.) If however, it’s suppressed due to a 4 hour hard race the day before, then yes you may want to make an adjustment. However in the latter case, Training Status would be “deep” yellow (i.e. if you’re projected to be yellow for a few days) and you’ll recognize you might need ride easier. Note that the easy riding isn’t because it’ll make you fitter directly. It just helps you recover faster so you can get back to doing more sooner.

3 Likes
#12

I’m coming to the conclusion that there are two aspects to this training thing. One - training for performance and Two - training for health and well being. There is obviously a large overlap between the two but I suspect that overlap gets smaller as you get older. The elite Pros are undeniably some of the fittest people on the planet (if indeed they are from our planet) but I think its arguable whether they are the healthiest. If you get the benefits of long rides below LT1 by staying as much as possible below it then something that potentially gives you a real time view of where you are as regards internal strain as opposed to external stress is something to investigate. I’ve seen a range of 165W-195W in my LT1 depending on how fatigued I was, how well I’ve slept, whether I’ve just eaten etc. My Xert LTP is bang in the middle. If I’m doing a workout at LTP the monitoring Alpha1 can indicate to me that I need to trim or increase the power on that particular day or indeed suggest I pull the plug completely as as happened on a couple of occasions when I felt OK prior to getting on the bike . Another benefit is that it gives me another number to look at when I’m on the turbo and helps keep me more engaged.

1 Like
#13

somewhat on topic but a side step, if one wants to assess DFA Alpha 1 the protocol that Bruce Rogers suggests and I think noted in @ManofSteele article in PEZ was doing step increments from say 60% TP up to a point where you cross the DFA a1 point of 0.75. I was thinking that this may not be all that precise becasue of the steps involved. Why not do a slow ramp, to cover the same time span as the steps but a gradual increase from say 60% to 90% over say 40 minutes and then watch DFA a1 as the stress/power increases? Does this make sense to others? I would hope to see a steady decline in the DFA a1 reported. I have a plan to try it out and see how it goes. Then as time passes and hopefully fitness improves one should see a gradual increase in DFA a1 0.75 vs power if in fact as literature indicates it is closely aligned with aerobic threshold.

1 Like
#14

The main idea is that you want steady 2-min windows with constant power to get an accurate representation of DFAa1, since alpha-1 requres a 2-min computational window.

#15

that is because of the short coming of the HRV logger calculation you only get a reading every 2 minutes, they have told me they intend to report data more frequently in future updates. Is hold time real needed, for instance mathematically wouldn’t a rolling average be equivalent since every 2 minutes is unique if off set by even a second… Bruce mentioned the 2 minute window is a shortcoming of the HRV logger and DFAa1 it reports he prefers other loggers and in particular Kubios. Other platforms report the data at a higher frequency (Runtastic or Fatmaxxer he mentioned). But I am curious why steps would be better than slope from a physiological point of view, because DFA a1 is a slope it has to have data points over time range to be calculated. By using a slope for power increments the change is more gradual and does not cause sudden changes in heart rate or maybe RR intervals. Another thought that comes to mind is that because heart rate tends to lag behind changes again a slope might be better? these are just questions not disagreement.

1 Like
#16

Do I have it right that the DFAa1 is used to establish your LTP and TP? It would follow that you would have to wait for you HR to stabilize after each step effort increase otherwise the curves would break at different points other then LTP and TP… potentially the step period could be shortened but a slow ramp rate can not be used (HR lag is hard to estimate).

Let me know if I have this wrong.

#17

to you as well as @ManofSteele the DFA a1 needs a span of data to be able to have enough for spectral analysis. Other programs can report DFA a1 every 5-20 seconds using a rolling 2 minute window calculation method. So there is nothing magic about 2 minute reporting intervals, that is just the unfortunate way HRVLogger does it. The 2 minute window is just a good rule of thumb to be able to get enough data to do the DFA analysis on. As for whether a ramp or steps is better is still a work in progress. There is no need for HR stabilization. HRV is based upon RR variation and this is a measure of the central nervous system stress. So as stress increases then the HRV goes down. That is why you do HRV readiness recording each morning on waking, the resting HR may be the same but the RR variability will change with stress. It is the latter you are measuring not the response of HR.

If you want a more frequent real time value for DFA a1 the only option I know of right now is Fatmaxxer but it is only available as an Android app. For me that does not work unfortunately. Kubios (free version) can report DFA a1 for any part of an RR series but it is not easy to see any graphics for the time series of DFA a1. Runalyze can report and present DFA a1 as a time series. I have yet to try it, though it looks like it will do the job of seeing the whole picture. So at this point the software is not quite there yet. I was told by the people at HRVlogger that they are thinking of adding more frequent reporting to the app in a future update… fingers crossed it is soon.

1 Like
#18

@ManofSteele @rgalisky a link to a blog post from Bruce Rogers. Muscle Oxygen Training: DFA a1 - Runalyze vs Kubios vs Logger results in a cyclist
of note see that Kubios, and Runalyze both are reporting 5 second DFA a1, as well as HRVlogger @ 2 min per reading output. Also note the power vs DFA a1 curves… the power is not stepped and the DFA a1 tracks along nicely in a linear manner in this case study. I still think there may be need for more data but I do believe that a slow steady ramp is better than 5 or 10 watt steps. Time and data will tell.

2 Likes
#19

Runalyze settings are configurable.

#20

yes and the site allows you to merge RR data from HRVLogger into a FIT file and then you can see DFA a1 at very high resolution. I am still working to understand the settings but it looks great and will work it seems for my use case. So instead of being held to one DFA a1 reading every 2 minutes you can see the DFA a1 in the range of every few seconds as well it automatically derives the aerobic threshold from the power vs DFA a1 graph. Very cool!