HRV-Guided Training

I asked Marco what did he think about HRM Pro comparing to Polar H10 and he said both are equivalent.

I’m comparing RMSSD and SDNN. I measured it simultaneously.The average HR and the graphic in both apps are the same. I imagined maybe they are treating artifacts differently.

I started to use Kubios because it has Baevsky Stress Index, ant it would complement my data.

What exactly did you do to measure simultaneously?
2 phones, 2 apps and 2 straps or 2 phones,2 apps, 1 strap or maybe even 1 phone, 2 apps, 1 strap?
And just to be 100% sure: these are the RAW rMSSD and SDNN values?
Trend looks very similar and that’s the most important, but nevertheless, I would expect the values to be almost identical…
One thing that differs in between the apps is the HR by which they switch from 0.05 (rest) to 0.20 (active) as artefact threshold. In your case probably not important at all because the chosen HR is somewhere in between 70 and 100. Your RHR is likely way lower when I see your HRV scores.
Kubios is not very informative when it comes to the phone app. Elite says they use the Kubios Premium method. Difficult to say where the differences are coming from.
Elite uses 2*Ln(rMSSD) as readiness and complements with subjective socres.

I’m using 1 phone, 2 apps, 1 strap and I do the measure simultaneously.

But I think the values are a little different. Some days they are identical and another very different. That’s why I think one of them are not correct, because I think the trend is different.

That’s my today’s morning measure. As shown, the average HR is identical (and always is the same). And the graphic seems to be the same, just inverted, because Kubios shows R-R Intervals and EliteHRV shows his own mathematic method.

So would that be the average EPOC for each activity?

Yes i think, so you can see for example average power/epoc over time 🤷🏼

I think its peak EPOC

I stopped using HRV4 and Elite and stuck with Kubios as I just couldn’t trust their subjective scores - there just seemed to be no transparency. That’s why I started using the intervals.icu ‘wellness’ page/s.
I also used the Kubios desktop app for a while but (as yet) I haven’t taken the time to (find out how to) transmit the kubios app info to it. Previously I exported from Elite/HRV4 and imported from there, but that was a pain.

A nice article from Marco (HRV4Training) explaining about overnight HRV:

The hrv4training score is a logarithmic conversion, you also have the option to see the rMSSD value not only this conversion.

This is a great thread! I am considering getting an H10 strap and doing morning measurements. From this thread, HRV4Training seems to be a popular option. I was wondering how you folks transfer the data from the app to intervals. Ideally, I would imagine measuring in the morning and then having the data automatically synced with intervals. Is this possible or does one have to do manual transfers? Do you recommend tracking other wellness metrics within the app or on intervals directly? Sorry if the question has been asked before and looking forward to some tips.

Hi.
After taking the morning measurement in “HRV4Training”, just select “Export to Dropbox” and everything will be sent to the cloud.
After pairing “Interval.ic” with “Dopbox” completely automatic, the first one will download data from Dropbox as soon as it is updated (and this is done by HRV4Training).

The process is as automated as possible, requiring only one-time setup before first use.

3 Likes

Look for the discussion about using Excel for the HRV, as it’s a really good way to track readiness.

I’m finding that I like graphs not scatterplots because I can better see trends over time. In trying to create graphs that provide insight…

Do I understand this correctly:
i. HRV above baseline + rHR below normal = PSNS overreaching, SNS exhaustion
• warning sign of exhaustion
• possible infection

ii. HRV below baseline + rHR above normal = PSNS fatigue, SNS heightened activty
• exercise overreaching
• infection

Noting the black lines and each day’s values below:
i. Is this period an example of (i)?

ii. Is this period an example of (ii)?

If the above is correct, and if the graphs are organised such that HRV is above and rHR is below, then:
i. the two Base lines further apart is a warning sign of (upcoming) exhaustion;
ii. the two Base lines closer together indicates an overreached state.

Yes, no, thoughts??

Yes, for trend indications the HRV4Training graph is better suited.
Some remarks when interpreting/discussing this chart:

  • Always use the correct terminology or it becomes very confusing
  • The bars are daily values and signal acute events
  • The baseline is a short term trend (7 day avg) and serves as your actual state. HRV is inherently fluctuating a lot and you can’t base anything besides acute events on daily values
  • The shaded zone are your normal values

i. SNS exhaustion is probably a bit too severe, I’d call it Acute fatigue. Should be easy to overcome with one or two rest days. I can’t tell anything meaningful regarding infection because depending on the source, interpretations differ. First screenshot indicates this situation, especially because the CV seems also high.
ii. Can be sickness or difficulties to recover depending on how much both parameters are out of normal. Second screenshot indicates a low CV which in normal cases is good but given the other indications, this is probably parasympathetic saturation.

The basics are rather simple (for evaluating what to do):

  • Baseline and daily value within Normal zone: all is good
  • Daily value outside Normal and Baseline within: Acute event, follow-up the next days to see if it stabilizes quickly.
  • Baseline outside Normal: Something more profound is going on and you should prioritize recovery.

Longer term trends can be seen in the shaded Normal zone. Rising HRV usually means better lifestyle conditions improving health. Most people see a circadian rhythm following the seasons when looking at data over multiple years.

There is one more thing that was added to HRV4Training and is not included in this chart. There is a middle term trend indication based on the slope of a trendline drawn on the values from the last 2 weeks. This provides extra info on a number of parameters to indicate Stable-Rising-Falling trends.

I’m copying in here below the notes I made while reading the HRV4Training blog. You will not find all of these interpretations on one page because they are scattered around the entire blog. I did some preliminary work with all this to improve chart interpretation but it has been lying there for about 6 months while I was busy with the iThlete chart. I still have the plan to take this up again but lack the time to do it in the near future.
My notes:

HRV Baseline 7d Ln RMSSD (in HRV4Training the double of Ln RMSSD gives results between 6 and 10)
RHR Baseline 7d
HRV CV (CV value targets are often based on Ln Rmssd in literature!!)
	Training Load
	Normal Values 60d

In a general way: stable is good

Low HRV CV is not always good
	Non-Functional Overreaching risk if combined with suppressed HRV. Less variation due to high stress.
	Good adaptation if combined with normal/higher HRV. Less variation means coping well.

Higher HRV CV means more jumping around of HRV and difficulties to maintain homeostasis (stability is good, remember)

Higher HRV + Lower RHR typically Coping Well
Lower HRV + Higher RHR typically Not Coping Well

Reduced HRV and RHR together can mean parasympathetic saturation

HRV reflects more Training Load than Cardiorepiratory Fitness. RHR better reflects Cardiorepiratory Fitness.
HRV is more a marker of current response to stress.
During a Taper week, HRV can be lightly reduced and leads to optimal competition form.

How to analyse daily score
Strong acute stressors have immediate impact (24-48hr). Detecting non-trivial fluctuations is key. Compare daily value to normal values for HRV and RHR and combine with subjective scores. Check recent trend (7d baseline) for better insight. Low daily score and baseline within normal is less of a problem than several low daily scores that take baseline out of normal.



How to analyse longer term trends? Parameters
	Ln RMSSD (HRV)
		Increase: coping well, improved fitness. Look at training phase to put HRV in context. HRV increase during high load phase means good response. Can be functional overreaching. Followed by deload, HRV should re-normalize
		Decrease: not always bad. Can be parasympathetic saturation or tapering. Look at RHR to decide. Decrease outside of tapering combined with stable or increased RHR means poor response to acute or chronic stressors
	RHR
		Increase: more fatigue if acute. Less fitness if chronic unless it is during taper.
		Decrease: coping well, better fitness especially for longer periods.
		Stable: ideal. Look out for seasonal patterns by comparing to Normal values
	CV (HRV)
		Decrease: combined with higher HRV and lower RHR is coping well. Combined with lower HRV probably non-functional overreaching.
		Increase: trouble adapting to training or acute non training stressor. Combined with reduced HRV, warning for inappropriate load.
	Normal Range
		When things stabilize after negative response, it's important to check if values return to normal and don't stay low. That to detect periods of long abnormal responses.

The different trends
	Coping well: HRV stable or increasing, RHR stable or decreasing and CV low within normal. Proceed as planned.
	Stable: normal fluctuation, no strong changes. Proceed as planned but keep an eye on acute stressors.
	Maladaptation: HRV decrease, RHR increase and CV increase. Poor response. Reduce training intensity.
	Accumulated fatigue: HRV decrease, RHR increase and CV decrease or below normal. Longer period of poor health/response. Prioritize Recovery.
	Saturation: HRV decrease, RHR decrease. Analyze in context of training program.
	
Goal is to detect non-trivial changes. Auto-detect by calculating the slope of the parameter over a 2 week window and only consider changes that are greater then 1 SD of the average.
type or paste code here
7 Likes

First off, thank you @MedTechCD for your extended response!
Can you clarify a few things for me?

If I use incorrect terminology please correct me.

I was asking more along the lines of general principles when HRV is high and rHR is low, but I appreciate you looking at this particular example. There’s defo not enough info there to make conclusions about infection. There’s certainly a warning sign of acute fatigue.

The high CV indicates ‘Poor response. Reduce training intensity.’ whether it be in combination with (HRV decrease + RHR increase) OR (HRV increase + RHR decrease)??

What is parasympathetic saturation?

So clear! Thank you.

Is a “middle term trend” a 14 day moving average of HRV? What info, specifically, does it provide?

Yeah, it’s difficult to find clear information in one spot. Most if it is too cursory or far too in depth to make head or tails of. I appreciate you sharing the salient points you’ve gleaned.

Is CV always 60d moving average? I thought I read that some people use 30d. There’s a huge difference between 7d (@david’s HRV chart), 30d and 60d.

…. as long as Baseline (7d) isn’t outside of the Normal range for either HRV or RHR, correct?

So I’m learning. What’s the A&P of this? Does it have to do with ‘parasympathetic saturation’, (whatever that is)?
EDIT: Answer found at Heart rate variability (HRV) during taper | by Marco Altini | Medium
• The reduction in training volume might elicit lowered blood plasma volume, and therefore decreased stroke volume, and in turn, increase heart rate and reduced HRV. This reduction in HRV reflecting reduced parasympathetic activity and increased sympathetic activity during tapering could be potentially linked to a better physiological state in the context of competition.
• As we taper and reduce load, resting heart rate increases, re-normalizing, from the acute fatigue state in which it is often suppressed. As such, HRV will also decrease a bit.

What I’m not seeing is HRV increase + rHR decrease of Base (7d) beyond Normal (60d) zone. How would that fit into the trends list? A type of Maladaption?

“Saturation”, i.e. parasympathetic saturation?

1 Like

In your post you mentioned above/below ‘baseline’ while baseline is no reference for this. Only Normal Values are reference.

A high CV always means ‘instability’ and you should try to avoid that. Low CV means stable situation.

It’s not an average. It’s a virtual plot of the values for 14days where you apply a linear trendline and you evaluate the basic trend. I haven’t gotten around doing the full interpretation but it helps defining a value trend on a shorter term then the normal values but more stable then the 7d baseline.

CV is always 7days. The 60 days is only for the Normal values. You’ve probably understood this wrong because my draft notes are somewhat messy…

This one is a remark for long term changes, not for short term.

You see, there is still a lot to do because the same trends for HRV/RHR can mean different things when considered acute, middle long or long term. That makes it really complex and it takes time to understand. It’s also sometimes confusing because authors of posts/blogs not always specifically mention the time window they are thinking about.
Marco Altini’s blogs and medium account + his twitter account where he regularly interchanges opinions with Alan Couzens and Stephen Seiler (and other equally very knowledgeable people) are really good sources for objective information with science based research.

5 Likes

I reached out to experts for clarification regarding elevated HRV + depressed rHR.

HRV4Training: Daniel Rowland & Marco Altini
Any large deviation from the normal range, even in the “right” direction, but happening acutely is not necessarily something good. We see a similar impact for some users when they see a particularly high HRV score after a race or big training effort; in this case it is often a large parasympathetic response as the body tries to cope with the stressors it faced. …In periods of extreme duress, if you’re seeing any changes in HR or HRV where they fall outside the normal range, then it’s most likely a sign that you need to reduce the load.

Andrew Flatt Ph.D.
Was this increase in HRV observed during intensive endurance training? If so, you may have been experiencing the parasympathetic form of overreaching. This is well documented. We even wrote a meta analysis on this.

2 Likes

I ran with @Olly_Thomas’s matrix and @MedTechCD’s summary notes. Added parasympathetic hyperactivity + visuals to reduce text.

5 Likes

Folks
How easy is it to pull wellness data into an excel sheet? Is there a quick tutorial one of you could give me or point me to on the internet? Tks
I only ask because I have excel set up to find rolling 7 day HRV, rHR and CV% + 60 day HRV + rHR and rather than manually input HRV to the file every day I’d prefer auto update.

Edit - Now that I think about it I could add in a sheet to the existing ithlete chart which pulls in the wellness data already.

Edit 2 - Still would like to know how to pull in that data from intervals if anyone could show me how. Everyday a learning day!!

I don’t really know excel a lot, but via the API you can get both wellness data as well as data about the workouts in JSON format to further play with it.

1 Like