Whoop's sleep tracking is getting a much-needed update. Here's what's changing.
Whoop has begun rolling out an update to its sleep tracking that improves stage analysis and awake recognition.
The brand detailed the new sleep algorithm in a blog post last week, though the enhancements did not actually take effect until yesterday (24 February).
Whoop says there are two key improvements users may notice. The update has improved sleep/wake detection by 3% and sleep staging accuracy for light, deep, REM, and awake time by 7%.
It arrives after two years of research and validation against the industry gold standard for sleep tracking, polysomnography (PSG).
This includes data from a Central Queensland University study that found the Whoop was 99.7% accurate in measuring heart rate and 99% in measuring heart rate variability. Interestingly, this data was also derived from the last-gen Whoop 3.0 tracker.
In that same study, it was also found that Whoop could only correctly identify multi-stage sleep/wake time with moderate agreement compared to PSG.
“WHOOP correctly identified 58% of PSG light sleep epochs, 62% of PSG deep sleep epochs, 66% of PSG REM sleep epochs, 56% of PSG wake epochs and 60% of all PSG epochs. The main sources of error were that WHOOP misclassified 28% of PSG wake as light sleep, 20% of PSG N1 or N2 as REM, 32% of PSG N3 as light sleep and 23% of PSG REM epochs as light sleep,” the study reads.
More than any other findings, this section is perhaps the reason why Whoop has been working in the background since the study was published in 2022 to deliver this update.
In the app, we also spotted that historical data will remain the same following the upgrade, while Whoop also notes there’s been no algorithm change for recovery, strain, respiratory rate, or resting heart rate metrics.
Wareable’s view
I’ve been a full-time Whoop 4.0 user for a couple of years, wearing the device primarily on my bicep as a reference point when testing other wearables. While I love some features, such as the Strength Trainer (extensively discussed here), I’ve never been blown away by the sleep tracking.
It’s not one of the worst, and the findings from the 2022 study detailed above support that. Yet, I still have to edit sleep data on one or two occasions a week because my Whoop incorrectly begins sleep tracking at 8-9 p.m. (despite plenty of movement after this) when I sit down to watch TV. This leaves me with little faith in the rest of the tracking.
Really, what makes Whoop’s sleep tracking useful (when it works as it should) is the insights into our recovery. That’s why it’s among our best sleep-tracking wearables to own.
I welcome this bump in accuracy, but I am also interested to see whether this helps Whoop catch up significantly to Apple and Oura in tracking sleep. I’ll compare data from this latest update to some PSG sleeps in the coming weeks, so stay tuned for that.
