Fitbit vs. Zeo vs. Sleep Cycle: Accuracy

I’ve been using all three for at least a month now, so I feel like I can give them a fair shake. Let’s do a quick recap of what each one does.

• Fitbit – Accelerometer, used both as a pedometer during the day and as a sleep monitor at night when strapped to your non-dominant wrist (I still don’t know why non-dominant, but I’m willing to go with it). Conventional sleep labs also make use of an accelerometer.
• Sleep Cycle – There’s an app for that! (Sorry, I just couldn’t resist). This app for the iPhone uses roughly the same tech as the Fitbit. Accelerometer based monitoring. You place your phone on your bed, near – but not under – your pillow. It tracks movement via the mattress.
• Zeo – Completely different tech. Instead of monitoring movement, the Zeo focuses on brain wave activity to track sleep phases. The monitor comes on a headband that you wear to bed. As tech goes, this is most akin to the EEGs used in sleep labs.

In a way, it’s kind of unfair to compare all of these monitors to each other. Fitbit and Sleep Cycle monitor motion, while the Zeo monitors brain activity. These are completely different things and provide you with completely different information. All, however, have the stated goal of improving sleep health.

Let’s compare the graphs from each for the night of November 28th:

None of them actually agree on how long I slept. Fitbit claims 5:59, Sleep Cycle claims 5:20, and Zeo claims 6:11. But let’s look closer. They all agree that I went to sleep around 6 am (don’t ask). So far, so good. They all agree that I slept fairly soundly until about 8 am. There’s a small spike of movement with the Fitbit, but we do move at least a little bit while we sleep, so that’s not particularly worrisome.

This is where it gets interesting:

There’s only one instance in which they gave significantly different data, but that may be the nature of motion vs brain activity measurement. They all agree that something major happened at 11 – it did. I got a text message that woke me and required a response. However, the very act of responding to the text message appears to have borked Sleep Cycle’s ability to record data.

What’s striking is how much they agree. These monitors come from three separate companies, with their own particular algorithms, software, and tech. And yet most of the time, they say the same thing.

It’s like when you ask your boyfriend if you look okay, and he says yes, but you aren’t sure, so then you ask your bestfriend, and she says yes, and then, for kicks, you ask your neighbor, and he agrees. If three independent credible sources all say it, odds are good that it’s true. I am impressed.

But now, it is bed time for me, and I’d best get to it.

To be continued…