The Doctor Who Coaches Athletes on Sleep
As the NBA and NHL playoffs start, a Harvard sleep specialist advises rest, not more practice, for championship teams.
On June 14, 2011, Dr. Charles Czeisler stood by the side of a small stage, listening as a colleague introduced him to a crowd of fellow researchers. Just as he prepared to ascend the steps, his cell phone vibrated in his pocket. He’s not sure why he answered it. Maybe he thought it was an emergency; maybe he just wanted to silence it. Either way, he took the call.
From the other end of the line he heard the plea of a team physician for the Boston Bruins, calling from Vancouver. Deadlocked with the hometown Canucks, the hockey team was one game away from its first Stanley Cup title in 39 years. Each team had won every game at home—to win the Cup, the Bruins would need to take away the home-ice advantage.
“What should we do? We need to win this one,” Czeisler recalls the physician asking. He thought it was strange timing, to call the night before Game Seven. He’d never spoken to the Bruins, and he didn’t have time to speak right then. He’d call him back after his lecture.
Czeisler is a tenured professor at Harvard Medical School and a go-to expert for professional sports teams from every major league. In the age of analytics-as-religion, teams are looking for every possible way to squeeze more skill out of elite athletes. They consult experts on everything from the number of minutes a player should be on the court to how many fourth down conversions they should attempt.
But Czeisler recommends something much simpler: more sleep. Professional athletes need the same thing as new parents.
The director of the Division of Sleep Medicine at Harvard, Czeisler is known around the National Basketball Association (NBA) as the Sleep Doctor. Jovial, he presents most of the research with a slight laugh, as if to say none of this should come as a surprise. It’s sleep. And yet, it’s so poorly understood. Beyond sports, he’s also consulted with NASA and the Secret Service.
Czeisler offers the generally accepted advice—naps before games or shifts, seven to nine hours each night. But when he addressed a group of analytics-crazed team officials, sportswriters, and entrepreneurs at the 2014 Sloan Sports Analytics Conference, he said something less expected: It’s the sleep after an event, a lesson, a game, that is most important.
“Interestingly, if you don’t sleep the night after training, then even if you sleep the next night or the next night, you never learn,” he told me in a follow-up interview.
In a 2011 review article on the literature related to sleep, his colleagues Richard Stickgold and Erin Wamsley explained, “During all stages of sleep, the mind and brain are working to process new memories, consolidating them into long-term storage and integrating recently acquired information with past experience.”
Offering a rudimentary explanation of circadian cycles, Czeisler explains that when we sleep, we repeatedly go through 90-to-120-minute cycles. Within those cycles, there are two main types of sleep: deep and rapid-eye-movement (REM). At first, the cycles are dominated by deep sleep, as our brains drain toxins that have built up throughout the day. The brain also flushes out excess synapses—the connections between neurons—that form during waking hours. In the process, the brain consolidates memories and rebuilds energy stores. As the night goes on, the balance shifts, with more active REM dominating the final cycles. Most of the dreams we remember occur during REM.
Solidifying knowledge requires both REM and deep sleep. The first aspect, which happens during deep sleep, is basically a rehearsal: The brain files away the facts, practices the moves learned that day. The second part is integrating those facts and lessons into existing knowledge. This happens during REM.
Stickgold had proven the importance of sleep in integrating new memories in a 2010 study that tested the memories of 60 people. Split into two groups, the subjects learned a set of words, some real and some made up, that sounded similar to common words—like “cathedruke,” similar to cathedral. The first group learned the words in the evening, took a test, slept for the night, and took another test. The tests involved recalling as many words as possible in three minutes—free recall—and recalling words when prompted with the first few sounds—cued recall.
Participants had 10 seconds to recall a word with cued recall. The second group learned the words in the morning, took a test, stayed awake for the day, and then took the second test. The subjects in the first group showed improved free and cued recall after sleeping the night; the subjects in the second group, who stayed awake between tests, showed no measurable improvements in free recall and were worse when asked to recall the words on cue.
Both groups were tested for a third time one week later. This time, the sleeping group showed no improvement from the previous week. The subjects in the second group, though, showed improved recall once they had had a chance to sleep, indicating that sleep is crucial to integrating new words into existing vocabulary.
Czeisler also loves to point to a study conducted at MIT in 2009. By studying the brains of mice as they navigated through a maze, and then as they slept afterwards, researchers were able to determine that the mice were replaying new experiences in their brains as they slept. This replay during sleep proved vital to forming long-term memories and remembering how to navigate the maze.
Fully integrating the day’s lessons requires seven to nine hours of sleep, Czeisler says. “If you lop off the last two hours of the night, you’re missing most of the REM time of sleep that you would have,” and therefore missing out on important steps in integrating the new memories.
For the teams he advises, sleep is never more important than in extended series, like the potential seven-game bouts that NBA and National Hockey League (NHL) teams are embarking on this week as the playoffs for both sports begin.
“You notice some teams seem to learn the moves of the other team, so they can actually counter their offense more effectively in later games,” he explains. “So that team that might’ve lost the first few games starts winning.”
When Czeisler started working with Portland’s basketball team, the Trail Blazers, in 2009, the players and coaches were rarely getting more than two or three hours of sleep between back-to-back games. When traveling, they were expected to just sleep on the plane. At home, they had early morning practices. So not only were they missing out on that crucial post-game sleep, they were just sleep-deprived in general. While many of the benefits of getting enough sleep—clearer thinking, more consistent metabolism—are well-established, the particular effects of sleep deprivation are less understood.
Asking athletes to play on minimal sleep is the same as asking them to “play with one hand tied behind their back,” Czeisler says. “It’s making them do something we know degrades their reaction time, their ability to take in their training, to get the most benefit out of it. They spend all this time practicing but never get to sleep.”
“Definitely, we know that sleep deprivation leads to depression, high blood pressure, weight gain, heart disease, and probably mortality,” Dr. Steven Feinsilver, the director of the Center for Sleep Medicine at Mt. Sinai School of Medicine in New York City, told The Atlantic in 2013. Additionally, a 2012 study detailed how sleep deprivation increases a person’s risk of developing Type 2 diabetes. Earlier studies have shown that reaction time nearly triples when a person pulls an all-nighter. Normally about a quarter of a second, it increases to 800 to 900 milliseconds. It’s about the same as the difference between being sober and being legally drunk. For elite athletes, emergency room doctors, and cab drivers, among others, losing that half of a second is costly.
As Czeisler detailed the detriments of sleep deprivation at the conference, one particular effect caught the attention of just about every guy in the packed room—the lucky ones in seats, the latecomers kneeling on the floor, the ESPN analyst sharing the stage: After just one week of sleeping five hours or less each night, a man’s testosterone levels drop as if he’s aged about 11 years.
In other words, a 22-year-old NBA rookie will have the testosterone levels of a 33-year-old veteran if he doesn’t sleep. Testosterone fuels muscle and impacts decision-making abilities. There’s no way a player is performing at his peak when testosterone is depleted.
These numbers come from a 2011 study published in the Journal of the American Medical Association. Previous studies, the authors explained, had shown that “the majority of the daily testosterone release in men occurs during sleep,” and “testosterone is critical in male sexual behavior and reproduction, but also has important beneficial effects on muscle mass and strength, adiposity, bone density, and vigor and well-being.”
To draw a connection between these previous findings, Rachel Leproult and Eve Van Cauter studied the testosterone levels of 10 healthy male students at the University of Chicago. After restricting their sleep to 4 hours and 48 minutes each night for one week, their daytime testosterone levels had decreased 10 to 15 percent. The levels were lowest between 2 p.m. and 10 p.m.—prime hours for working, whether you make a living playing in the NBA or running a small business. By comparison, when men are well rested and aging normally, their testosterone levels drop about 1 to 2 percent each year.
How quickly testosterone levels regenerate after sleep loss is still in question. A 2012 study showed that younger rats bounced back much faster than elderly rats. Testosterone stores started to regenerate in the first day after sleep recovery for the 12-week-old rats; 20-week-old rats had to wait five days for the levels to start restoring.
Brain scans show that when a person is up all night, the connections between the prefrontal cortex—the area of the brain responsible for judgment—and the amygdala—responsible for emotion—are decreased. Overtired, the amygdala is 60 percent more reactive, according to a 2007 study.
“So now you have the emotional brain and it’s going off on its own without being kept in check by the judgment area of the brain,” Czeisler says.
Looking at this, it’s no wonder the number of technical fouls increases after back-to-back road games, he adds.
If there is a benefit to lowered testosterone from sleep loss, it could be lowered aggression. A February 2013 study conducted at Brock University in Ontario determined that when men are sleep-deprived, they are less likely to react to something aggressively. Sleep deprivation had no impact on female aggression.
In the NBA, performance in general plummets in back-to-backs, possibly because players are playing more emotionally than strategically; possibly because decreased testosterone levels are hindering their game. In Major League Baseball, home teams score 1.24 more runs when the visiting team has flown in on a red eye, according to a 1995 study. There’s minimal impact when the visiting team flies west, indicating that the sleep lost on a red eye is the determining factor.
Sleep research has helped countless elite athletes perform at their peak, but athletes are clearly not the only sleep-deprived group of workers. According to Van Cauter, at least 15 percent of U.S. workers get less than five hours of sleep each night. Czeisler and others have said that we live in a chronically sleep-deprived society.
“Today, 30 percent of all employed U.S. adults and 44 percent of night workers report averaging less than 6 hours sleep per night, whereas 50 years ago less than 3 percent of the U.S. adult population slept so little,” he wrote in a 2013 Nature article.
In a 2006 interview with Harvard Business Review, Czeisler advised that companies should not expect workers to log more than 16 hours in a row, or to drive or work after an overnight flight. "We now know that 24 hours without sleep, or a week of sleeping four or five hours a night induces an impairment equivalent to a blood alcohol level of .1 percent,” he said. “We would never say, 'This person is a great worker! He's drunk all the time!' yet we continue to celebrate people who sacrifice sleep for work."
The culture of overworking employees was brought into question in August 2013 after a 21-year-old intern at Bank of America’s London office died suddenly. He had worked about 72 hours without sleep leading up to his death. An autopsy later ruled it an epileptic seizure—which can be brought on by exhaustion, as The New Yorker's Ruth Margalit explained—but the tragedy led many Wall Street firms to at least review workloads.
Long before reaching the workplace, there are countless instances when sleep deprivation impacts performance. College exam season is quickly approaching. Many of the students will sit through a review session, head home, and cram over the material all night. If they do, they’ll be making a huge mistake. They’re less likely to retain any of the information learned during that review session, and they’ll be exhausted when they sit for the exam. The same goes for lawyers prepping for a case or traders burning the midnight oil to get an edge.
“Sleep plays such a big role in learning and memory,” Czeisler says. “It’s a simple way to have an edge and be able to perform at your best.”
When he works with a team, Czeisler’s first move is often encouraging teams to cancel early morning flights and practices. The old-school approach of toughing it out, attempting to beat jet lag, he says, is completely bogus, not to mention counterproductive.
That’s what he told the Bruins’ physician on the eve of Game Seven of the 2011 Stanley Cup Finals. Calling back after his lecture, he asked about the team’s plans for the next day, how they would spend the hours leading up to the 5 p.m. puck drop.
They had a practice scheduled at 10:30 a.m., he was told. That’s when he made his first suggestion: cancel the practice. The guys need to take a nap instead. It would be the equivalent of a 1:30p.m. nap in Boston, and would help them make up some of the sleep they’d lost while traveling.
The front office listened. Instead of practicing, the players napped. And that night, the Bruins handily captured the title, winning 4-0 and hoisting the Cup for the first time since 1972. Maybe the nap made the difference; maybe the Vancouver Canucks were more sleep-deprived. Either way, by the time a team makes it to Game Seven, after playing more than 100 games together in the regular season, there are few things it can change to see an immediate impact.
A call to the Sleep Doctor, though, could make all the difference.