TST Whoop Group

[TN94z]

No Bluetooth problems here either.  My only issue lately is that I did not wear it much during my 2 weeks off...lol. It started off that I forgot to put it on one day as I was rushing out of the house.  Then it spiraled out of control. My work schedule is back to normal now so my "routine" should be back to normal.

[iacas]

In less good news, 20% today with an HRV of like 39. Woof.

😄

[saevel25]

Just looking at the past 15 days worth of data. 

Recovery	HRV	RHR	Sleep	Sleep%	Resp	Strain
93	90	53	7.22	93	13.2	9.8
69	73	52	8.50	100	12.8	10
70	74	52	6.67	87	13.1	4.2
61	69	55	8.38	100	13.3	4.6
56	69	56	8.72	100	13.5	7.5
71	75	52	8.10	100	13.0	6.8
90	84	51	8.78	100	13.6	5.6
97	95	48	6.83	91	13.0	4.2
96	94	51	8.67	100	13.2	4.1
98	103	50	8.67	100	13.3	4.8
91	96	52	8.13	100	13.2	8.7
71	88	52	8.15	100	13.0	5.2
74	89	50	7.37	97	13.1	5.3
64	86	49	7.60	99	13.0	4.6
55	80	52	7.00	92	13.2	3.9

I am guessing my 55 Recovery score is because I have three days in a row of under 100% sleep. My HRV is right around my baseline (70-88). At the top, I had 2 out of 3 days around 90% sleep and the next two days were my worst recovery. As soon as I got enough sleep over the next 2-3 days, then it jumped back up into the 90's. It looks like there is some lag. 

[iacas]

35 minutes ago, saevel25 said:

Just looking at the past 15 days worth of data. 

Recovery	HRV	RHR	Sleep	Sleep%	Resp	Strain
93	90	53	7.22	93	13.2	9.8
69	73	52	8.50	100	12.8	10
70	74	52	6.67	87	13.1	4.2
61	69	55	8.38	100	13.3	4.6
56	69	56	8.72	100	13.5	7.5
71	75	52	8.10	100	13.0	6.8
90	84	51	8.78	100	13.6	5.6
97	95	48	6.83	91	13.0	4.2
96	94	51	8.67	100	13.2	4.1
98	103	50	8.67	100	13.3	4.8
91	96	52	8.13	100	13.2	8.7
71	88	52	8.15	100	13.0	5.2
74	89	50	7.37	97	13.1	5.3
64	86	49	7.60	99	13.0	4.6
55	80	52	7.00	92	13.2	3.9

I am guessing my 55 Recovery score is because I have three days in a row of under 100% sleep. My HRV is right around my baseline (70-88). At the top, I had 2 out of 3 days around 90% sleep and the next two days were my worst recovery. As soon as I got enough sleep over the next 2-3 days, then it jumped back up into the 90's. It looks like there is some lag. 

Is that on the Whoop site? I only get charts, not a nice table like that.

Stuff like this is nice, but where is the data copied from?

[saevel25]

1 minute ago, iacas said:

Is that on the Whoop site? I only get charts, not a nice table like that.

Stuff like this is nice, but where is the data copied from?

I just typed it into excel. I was curious to see what I can make out of the data. I know I’ve been lazy with not working out hard. So, one variable was set there. 

[iacas]

33 minutes ago, saevel25 said:

I just typed it into excel.

Oh. Okay. 🙂

[iacas]

I read this just now. (It may have been posted before, and since apparently Reddit embeds are full-size, it's at the bottom).

Here's what happened to me this morning. At about 10:30pm last night, I went to watch the second part of the Tiger documentary. My wife fell asleep pretty quickly and my kid watched the whole thing with me. I turned it off and fell asleep pretty quickly, around 12:07am.

My recovery was 75% and my HRV around 85 the next morning, but my "sleep" time was 11:03 or something. When I adjusted that to 12:07, my recovery dropped to 51% and my HRV to 57 or something.

HRV is supposed to be calculated in the last 5 minutes or something of your deep sleep, so I'm not sure why adjusting the start time - while still doing the same "resting" as I was doing, by just lying in bed - should have any effect on my HRV or my recovery. If my HRV is reported as a different number I get why the recovery would change, but why did the HRV change when I left the wakeup time as 6:59am?

 

 

[iacas]

Quote

 

I apologize for any misunderstanding here. When you manually edit your sleep on WHOOP, we re-process it and take into account the new time window that you provided. WHOOP measures HRV and RHR during the last 5 minutes of the last slow-wave sleep cycle to calculate your Recovery score, so changing this period can potentially change your HRV, RHR, and Recovery.

If a sleep is edited, whether by minutes or hours, the reprocessing of the whole sleep occurs and WHOOP processes sleep second by second. This 'tells' WHOOP that the first sleep was not correct and, so it believes the first place that RHR and HRV were taken was not correct either. Then the algorithm will skip over that first measurement point and completely pull from the next most accurate SWS portion of the night.

HRV is a highly sensitive metric and it will fluctuate throughout the day and night drastically based on the state of the autonomic nervous system. So, it is normal to see this change in the metrics.

 

This doesn't really answer my question.

[bweiss711]

Just spit-balling here.. but what I think must be going on is that whoop is calculating SWS cycles against some sort of baseline.  If the baseline calculation was affected by shifting the start of the sleep period, perhaps that affects when and where it calculates the subsequent SWS cycles.  That's just a guess at trying to make sense of it, though.

[iacas]

1 hour ago, bweiss711 said:

Just spit-balling here.. but what I think must be going on is that whoop is calculating SWS cycles against some sort of baseline.  If the baseline calculation was affected by shifting the start of the sleep period, perhaps that affects when and where it calculates the subsequent SWS cycles.  That's just a guess at trying to make sense of it, though.

Yes, maybe it doesn't really know when you're in SWS, and it's kinda "guessing" at it via some algorithm or something.

[iacas]

A better answer:

Quote

That is a great question and I am glad you followed up with us about this. To further explain, WHOOP typically measures HRV and RHR during the final 5 minutes of the last slow-wave sleep (SWS) cycle. There are exceptions because sleep stages aren't purely discrete categories, but continuous distributions where there is deeper light sleep, lighter deep sleep, and extremely deep sleep, etc.

For clarity and consistency, WHOOP uses the standardized definition of these stages and applies a threshold to break down the continuum. Behind the scenes, we have a continuous distribution of these stages where every second of your sleep is assigned a "degree of slow-wave sleep." So when a sleep is edited, and the app completely recomputes the entire sleep and all the sleep stages, there may be periods of the sleep that have been designated SWS that weren't designated that way in the auto-detection. This is why you are seeing this difference in the HRV measurement. 

 

[bweiss711]

9 minutes ago, iacas said:

Yes, maybe it doesn't really know when you're in SWS, and it's kinda "guessing" at it via some algorithm or something.

Your situation still calls into question the accuracy of or the confidence we should have in the recovery/hrv numbers.  In reality, your SWS cycle occurred when it occurred.  And thus, your HRV in the last 5 minutes of that cycle was what it was.  However, if the change in the algorithm can change the estimate of when that was and have it result in a 25% change, then what does it really mean?

I've had a similar debate going on about my numbers.  I actually manually start my sleep activity when i lie down to go to sleep.  Last night, that was 10:15 or so, and I know i didnt actually fall asleep until close to 11.  However, my whoop data counts 10:15 to 11:00 as light sleep with only one occurrence of "awake".  Does that mean lying still in bed while awake isn't any different than light sleep?  If so, thats fine.  If not, then the same thing holds where the baseline calc could be off, and that changes the rest of the metrics.

3 minutes ago, iacas said:

A better answer:

Yes, that is a better answer.  Still uncomfortable with a 25% difference in recovery.

[iacas]

27 minutes ago, bweiss711 said:

Your situation still calls into question the accuracy of or the confidence we should have in the recovery/hrv numbers.  In reality, your SWS cycle occurred when it occurred.  And thus, your HRV in the last 5 minutes of that cycle was what it was.  However, if the change in the algorithm can change the estimate of when that was and have it result in a 25% change, then what does it really mean?

I'm fully on board with you, yeah.

I've had my Whoop auto-detect a "nap" when I sit and watch a movie mid-afternoon. I had one of those around Christmas when we watched a movie. 🙂

Of course, I don't expect it to know if I'm actually asleep or just lying there not moving much. I'm sure my breathing slows a bit, my heart rate drops a bit, and it's a lot like "light sleep." May even be almost as restorative.

So I don't have a problem with that (though I did delete the nap, as I wasn't actually asleep).

[iacas]

BTW I was wrong. It was 85 to 67. I had though it went to 57.

Here are the images:

What you'll note is that the areas marked as SWS are the same. So…

And

[bweiss711]

23 minutes ago, iacas said:

 So I don't have a problem with that (though I did delete the nap, as I wasn't actually asleep).

Same.  Usually happens on Sundays during football.  I delete as well, but mostly because it affects the recommended sleep time.  And therefore throws off the performance metric and bedtime suggestion.

But if sitting and doing nothing for 90 minutes is the same as a nap, then there's nothing to worry about.  But I dont think it is.

[iacas]

I’ve had some shitty recoveries lately.

These are the day after, so “Tuesday” is my sleep “Monday night”:

• 25% on Tuesday (5:57, 11:51 to 6:20 in bed)
• 27% on Wednesday (6:47, 11:46 to 6:52)
• 23% on Thursday (6:45, 11:10 to 7:02)^
• 32% today (5:22, 10:44 to 4:42)

^ I had a bit of a headache yesterday and was slightly congested. I took a single Aleve at about 4pm, and went to bed a bit early… and then woke up at 4:20, tried to fall asleep for 20 minutes… and gave up and got out of bed. Got a lot of work done between 5am and 7:30.

I felt “fine” all the days. I could have played soccer or walked 18 holes. Still, I do wonder if there was something “off” and the mild headache was interesting (not so mild I didn’t take anything, but it wasn’t a migraine).

P.S. My sleeping respiratory rate is normally about 16.2. Tuesday was 15.3, Weds and Thu 15.9, today back to 16.1.

[saevel25]

There has been some historical review of sleeping patterns. They noted that it was very common for people to sleep 4 hours, wake up for while, then go back asleep for another 4 hours or so. At a certain time this started to become uncommon to do and that you need to sleep a full 8 hours in one go. 

I am the type of person that just gets a crazy amount of deep sleep. To the point that I was sleeping through a fire alarm in college, and my roommate had to wake me up. 

Here is my recovery, 

Tuesday: 56%
Wednesday: 73%
Thursday: 98%
Friday: 75%

Strange data is last weekend with my Strain at 10+. I didn't do anything at all to warrant that. Weird. 

[iacas]

Maybe you were a little sick, and your HR and breathing rate were up?