Sleep trackers

My friend just asked me this and I’m embarrassingly ill-informed but I think many of you have high-quality opinions on this and it would be great to collect those here!

I wondered if you could give me your opinions about sleep trackers. I’m currently considering:

An Oura, but: they’re expensive, [spouse] and I can’t both gather data from one cause they’re sized, and I’m not 100% sure I’ll be able to wear a ring.

A Garmin Vivosmart, because they’re cheap and not owned by Google, and they have all of the sensors that anyone might need, but I’m skeptical that they are actually focusing on doing a good job of data analysis, and not just a play to be in the space.

A Fitbit, because duh, but I’m not sure about the Google acquisition, and also they don’t have their spo2 sensor hooked up yet, and seem to be saying they’ll be requiring a subscription to get the data out of that.

Am I missing a better option? Or misunderstanding something about these?

1 Like

oura was nice. wearing a ring is real simple and way better in lots of ways than other fitness trackers. and i think sleep tracking is where oura really shines.

there are ppl making sensors that go on the bed. e.g. https://www.beddit.com/ – not sure top-of-head who else is hitting this side of the market, and i’ve no idea if they’re any good, have api, etc, etc.

1 Like

If they have an android smartphone I would recommend Sleep as Android which can be used as a sleep tracker either on the bed or using sonar on a side table near the bed plus integration with Beeminder.

1 Like

On iOS there is an app called “SleepScore”, also using your speaker and microphone as a makeshift sonar. It tracks your sleep phases and even your breathing rate and got a smart alarm that works really well.
On my phone (iPhone 6s) I can not hear the sonar and I got very sensitive hearing. And I was told they do calibrate the sonar on each iOS device to get the most accurate reading.
I’ve been using it for a couple of weeks now and the results make a lot of sense to me.
It also integrates with Apple Health to track your time in bed and your time asleep.

https://apps.apple.com/us/app/sleepscore/id1364781299

I have previously used a Withings Pulse HR and it’s a piece of garbage. Don’t bother with it.

Has anyone got actionable information outside of “you spent X hours in bed” from these? What do you do differently based on the information you are capturing?

2 Likes

a) For me it is part of my personal “what contributes to my chronic tension headaches” research. I keep a log of my headaches and other things like my water intake (and how much I slept) and once I acquired enough data I’m gonna continue exploring the dataset for correlations. I might even throw some machine learning onto it.

b) Especially SleepScore gives you a ton more metrics than just how long you slept. Here’s some info about last night:


SleepScore also boils all of these metrics down into a single SleepScore metric and in my experience it reflected rather well how well rested I felt the next day:

There is actually a number of studies that went into the making of the app and I tried to track them down in the past but ran into a bunch of broken links.

1 Like

I’ve used a couple of iPhone apps and Apple Watch apps, currently using AutoSleep on the watch. It gives me stats similar to the ones phi shows for their app. I find some of them useful (the general sleep ratings match my feelings about how good a night I had; the sleep credit/debt percentage match how tired/energised I am through the day), some of them garbage (it gives a “readiness” rating when you wake up which bears no resemblance to how ready to get up I feel), and some of them are interesting data I can act on (consistency of bed/waketime). I’d like it to tell me more info about night time wakings but although there’s a sleep efficiency rating (% of time in bed you were actually asleep) it’s no actual use because I tend to give up when I can’t get back to sleep so I get 100% efficient for sleeping between 23:00 and 04:00 when really I’d rather stay asleep until 07:00.

I too have been searching for the magic correlations to get rid of headaches, I always feel that sleeping is most of the problem, but Exist’s correlations show me that lack of water the day before is actually better correlated to headaches. I think this might be down to the effects of lack of sleep being over multiple nights though which Exist doesn’t currently correlate against. I still think trying to sleep better has to be helpful.

My problem is that I have to keep a close eye on the data; because you’re wearing it in the day too it picks up naps well, but since I’m a restless sleeper it often thinks I’m awake when I’m asleep and also asleep when I’m awake (on the sofa with my laptop for example). It has calibration to deal with this and uses heart rate dips as well as movement but there’s only so much it can do for someone who sometimes moves more while asleep than when awake!

As far as actionable data go I’m trying to work on getting back to sleep quicker (or at all) when I wake up in the night and getting to bed in time to have a chance of staying in credit with the sleep bank. The second might be something I could beemind though I’m not doing currently, the first not so much.

2 Likes

thanks for mentioning Exist. I had no idea that this is a thing. I already got a python notebook started which massages my data into pretty graphs. Definitely something I got to evaluate!

I absolutely love Sleep Cycle on iOS (https://www.sleepcycle.com/), and have been paying for a subscription to it for about 2 years or so and intend to continue indefinitely.

It does a great job of figuring out when I’m asleep, with a few exceptions when I’m lying in bed for a long, long time unable to sleep, but not moving much at all. (That happens about every two months.) I like the stats I get from it, and I find the alarms really pleasant to wake up to. I’ve tried a few other sleep apps (I can’t remember their names, because I deleted them after just a day or two) and ended up immediately coming back to this one. The only downside (and the reason I tried other apps) is that if you share a bed with someone who snores, it’s probably going to track it as you snoring. Since I’ve historically snored under a minute per month, I just ignore that stat, but if that’s an important one to you and you sleep with someone else who snores, it might be a deal-breaker.

4 Likes

Seconded. I’ve been using it for years and it keeps improving. Haven’t gone Pro, but am Pro on their other app, LifeCycle.

Their alarms are a killer feature. 1. they try to go off when you’re already kind of awake 2. they start quiet and increase in volume 3. it’s a randomly selected sound, so I don’t dream it prematurely…

4 Likes

Has anyone set up Beeminding SleepCycle? I’d want it to pull automatically - does that work? Ideally to beemind wake-up time.

I’m not a premium SleepCycle user, but would consider it if needed. iOS, if it makes a difference.

I’m beeminding hours of sleep via Fitbit now, and I’ve tried using stickK for commuting to using SleepCycle before (never derailed, but found the manual logging tedious). I tried beeminding hours in bed with Apple Health, but the numbers seemed wildly high compared to actual time in bed (might be logging stationary phone time as sleep?)

I agree, @philip - their alarms are the best feature :star_struck:

2 Likes

I used it for a long time, tracking both the in-bed time and the asleep time by connecting it to Apple Health and then connecting Beeminder to Apple Health. Rarely, Beeminder wouldn’t be able to pull the sleep data from Apple Health but most of the time, that was smooth as silk for me.

1 Like

(Oh, lol, I just scrolled up and see that I already said things in this thread. Welp… hehe) ¯\_(ツ)_/¯

1 Like

My Apple Health only shows me sleep stats from SleepCycle. Don’t know whether that would help your case.

Thanks @philip and @mary - I’m pretty sure my phone was pulling stationary time, since it had hours in bed even when I wasn’t using SleepCycle, but I’m play around with settings and permissions and see if I can get a commitment to pull reasonable stats for’ time in bed’ from SleepCycle via Apple Health. I feel like time in bed is more directly under my control than time asleep, so that would likely be the better metric for me to mind.

I do really want to mind wake-up time in some automated way. I’m sure I read a help or forum article about tracking wake-up time but haven’t been able to find it again through searches (or maybe I mis-remembered how it was supposed to work), since nothing I’m finding now seems like quite the right fit. Maybe I can get Beeminder to pull wake-up time from Sleep Cycle.

I’m sure you’re right. I haven’t turned on the Apple Health sleep stuff on my phone, which makes SleepCycle the only source for me. If you don’t value whatever else iOS is doing for your sleep, maybe there’s a way of turning it off or excluding it from the data set.

1 Like

Great idea, @philip - I’ll try that! I have a Fitbit tracking actual sleep better that Apple Health or Sleep Cycle already, so it’s worth trying! Maybe I can get Beeminder to pull wake-up time and finally get the automated waking up accountability I’ve been trying for :ok_hand:

@mary, @Philip @adamwolf - I found the issue (and solved it!!!). Saving here for posterity and future reference.

Once I found the more detailed iOS Apple Health sleep log data, I found that it WAS pulling stationary time and adding it to the sleep cycle time. Of course the two weren’t identical, but it was close to doubling my hours in bed (no WAY I was in bed 10hs… more like 4, hence the goal! :woozy_face:).

I could routinely manually delete the iPhone duplicates, but as you might have guessed by now, I don’t like that kind of inefficiency if I can help it, so I dug around. I explored how to delete and remove sources, remove permissions, etc. I googled and still no luck. The problem was, my phone wasn’t listed as a ‘source’ of data, so how could I remove it?!?

It had to be somewhere though, since the data points were showing up as ‘Luna’s iPhone’ - clearly iOS was generating these data points. Turns out it’s a ‘Track time in bed with iPhone’ SETTING under options in Apple Health sleep :roll_eyes: See highlight in image.

Not very intuitive, but it’s a simple toggle off now that I know it’s there. I haven’t slept yet, but I’m 99% confident THIS was where the mystery data was coming from. Could work nicely for someone who doesn’t use Sleep Cycle/another tracker (or leave their phone idle for hours when they aren’t sleeping…) and wants to set up a goal to track time in bed.

Thanks again Mary & Philip re: tips to set up SleepCycle. It’s working beautifully AND forcing me to use this alarm that’s so wickedly effective at waking me up on time (it KNOWS!). I used to have a manual stickK commitment for using Sleep Cycle, and now it’s a two-for-one get more sleep/ time in bed AND set alarm automated goal that will sting if I’m not diligent. Perfect!! I love Beeminder (when I can get them set up how I envision them :stuck_out_tongue:)

3 Likes

Thanks for reporting back!

I had the same thing happening with stuff like run data or number of steps, where the data from an external device was being added to the data collected by my phone, without the Health app paying attention to when they were doubling up.

2 Likes

Nice! Thanks for tracking that down and posting it here! I’m sure that’ll help someone else in the future when they have the same problem!!

2 Likes