SDRR (for HRV)

My question wasn't as focused as it could or should have been.  Why I want to know is irrelevant.  Apologies. Better distillation:

 - What does the SDRR metric contribute to understanding a person's heart rate variability beyond the information RMSSD provides?
 - Does SDRR convey a level of confidence in the useability of that specific instance of RMSSD?
 - Is SDRR offered for people who believe it to be a better metric than RMSSD?

i find these values all quite volatile. I do a health snapshot and one more right away... and the HRV varies between 40 and 80... does this help? to me not quite... unfortunately all of this seems to be still not to be precise enough, even for our amateur level of expectation in such a device

I'm pretty sure I agree.  I've had quite a few times when a health snapshot provided no HRV info, and another snapshot immediately after that provided a reasonable, fairly believable RMSSD value.  The watch yells at me that my stress is high, my sleep poor, and my body battery often low - all allegedly based in some part on HRV.

On top of all that, "Athlete's Heart" (sinus arrythmia) seems like it should interfere with the algorithm, and a cardiologist told me that I have that and it's not something to worry about or to treat.  Seems like many if not most people wearing a sports watch will have the "Athlete's Heart" condition in addition to "Athlete's Bradycardia" (slow resting pulse), both almost certainly messing with a watch attempting to analyze an ECG taken from a wrist sensor.

I don't want to ignore my watch's assessment of HRV entirely, but it's not at all obvious the extent to which we should pay attention.

You may want to have a look at firstbeats website. They did all the algorithms and have plenty of documents explaining them. Maybe they are covering your questions too?

Excellent suggestion.  I have not yet found what I'm looking for, but they do seem to provide a fair amount of this kind of information.  Among findings in the first few minutes of searching is a document at Heart rate variability (escardio.org) which is pretty dry science-speak (to be expected).  First read of that suggests RMSSD is a shorter time horizon metric, while SDRR attempts to characterize variability for a longer time window (like days or weeks?).  Don't quote me; I might not have sufficient coffee today to be reading this kind of literature.  :-)

To follow your wording: I never had enough coffee to understand all their studies ;-) but they made me confident to trust that they know what they are doing. The folks are even pretty responsive in answering emails. So you may want to give this a try if you really would like to understand all bits and pieces. 
By the way, they also have nice recorded webinars. 

I wish I could take back my speculation about time horizons in the preceding.  This is how crappy internet information is born.  I looked at the RMSSD and SDRR metrics for six months of my Health Snapshots, and I see no discernable pattern. Truly wild swings from over 200 ms to under 20 ms within 24 hour periods. If SDRR had value as a longer-time horizon metric, there wouldn't be that kind of violent variability in such a short time.  What SDRR tells us continues to be a mystery to me.  Hopefully someday a cardiologist will see this thread and supply the answer.