Most people are familiar with the importance of getting about eight hours of sleep each night. However, most people are not familiar with the importance of getting those hours of sleep with at least somewhat consistent timing. WHOOP recently introduced Weekly Performance Assessments, and in them, a measure of how consistent your day-to-day timing of sleep is. Here we discuss some of the research that went into developing this feature and the details behind its calculation.
A 2017 paper out of Harvard University introduced a new metric called “Sleep Regularity.” The paper reported a significant positive correlation between this novel metric and academic performance (as measured by GPA) in healthy college students. In fact, students in the study with regular sleep and wake times had an average GPA of 3.72 while those students deemed irregular sleepers averaged only 3.24 on the 4-point scale; that’s the difference between an A and a B+ average.
The most exciting part of this finding, to quote the study’s lead, Dr. Andrew Phillips, was: “Sleep regularity was uncorrelated with sleep duration, suggesting that regularity captures another informative dimension of sleep.” To put this simply, if every night you get 100% of the sleep WHOOP recommends, you could still be missing out on potential sleep-related benefits if you are getting this sleep at inconsistent times.
Taking a moment to think logically, this isn’t that surprising. For example, we tend to feel better if we hydrate at somewhat regular intervals than we would if we were dehydrated all week but made up for it on the weekend with surplus hydration; why should sleep be any different? We therefore set out to quantify the relationship between sleep consistency and WHOOP data in order to understand why our members ought to be striving to improve the consistency of their sleep timing.
Before we dive into our exciting findings, let’s first back up one step and explain how we calculate sleep consistency.
WHOOP Sleep Consistency measures the likelihood a user’s state (sleep, or wake) is the same at a particular clock time on each of four consecutive days. We note that this is a little different from the Harvard Sleep Regularity Index because their index was developed as a summary statistic, looking at months of data at a time, while the WHOOP metric looks at day-to-day changes in order to provide actionable daily feedback.
The example data below, taken from May 20th-27th, will help us better understand this metric.
This horizontal axis of this graph shows the 12-hour clock time in the user’s local timezone. Each row shows a different date, ordered consecutively from the 20th through the 27th of May, 2018. The green bars show when the user was asleep and the remaining white spaces indicate wake.
Notice that on the weekend days, May 20th, 26th, and 27th (indicated with a slightly darker green) the user did what so many of us do, and slept in far longer than he had been during the week. While sleeping in is great for catching up on sleep debt, it can be disruptive to our circadian rhythms (biological clocks). According to Phillips, the disruption to circadian rhythms is the most likely culprit behind the impaired academic performance observed in his study.
So even though this user was averaging about 8 hours in bed per night, he would have been better off getting a consistent 8 hours every night than what he did – 7 hours Monday-Friday and as much as 12 hours per night on the weekend. In this case, his data translates to an overall mediocre sleep analysis in his Performance Assessment, reproduced below.
According to researchers at both Harvard and Stanford, the reason why getting 8 hours of sleep each night is not the same as averaging 8 hours per night is that the brain gets “thrown off” – an effect equivalent to always being jet-lagged.
Disrupting the circadian rhythm in this way prevents us from producing the sleep hormone melatonin. Since melatonin concentrations are highly correlated with sleep quality, WHOOP set out to measure the relationship between various measures of sleep quality and our new measure of sleep consistency. The results are reviewed below.
As it turns out, WHOOP Members who think they are making up for insufficient sleep during the week with extra sleep on weekends are often fooling themselves. Those whose average sleep consistency was in the Poor zone averaged 4 percentage points lower Sleep Performance (the percentage of total sleep each night compared to sleep needed) than did users in the Sufficient and Maximal zones, corresponding to average Sleep Performance scores of 70.6 and 74.1, respectively. While that may not sound like a huge difference, it adds up to about 2.5 more hours of sleep over the course of a week.
Members ask us all the time how to increase their slow wave (deep) sleep and REM sleep. While not the biggest game changer, those with greater sleep consistency averaged slightly more slow wave and REM sleep, and had slightly less time awake, than did members with the same amount of sleep time but lower measures of consistency. This means that consistent sleepers aren’t just getting more sleep, they are getting more restorative and efficient sleep than are people getting similar amounts of sleep at inconsistent times.
Lastly, our members feel the difference. Those whose average sleep consistencies were in the Sufficient and Maximal zones were 4% more likely than those in the Poor zones to report feeling rested or energized on our daily Recovery survey. Members whose average sleep consistency was Poor were also more than twice as likely to report feeling exhausted than those whose average sleep consistencies were in the Sufficient and Moderate zones.
Sound too good to be true? Why not see for yourself. Challenge yourself to get 7 days of Maximal Sleep Consistency next week!