According to the article, Fitbit One calculates sleep efficiency this way: time asleep / (total time in bed - time to fall asleep). But that formula doesn't agree with what Fitbit reports. Example: last night Fitbit reports a sleep efficiency of 75%. But calculate it: I was asleep for 7:19, in bed for 10:41, and took :31 to fall asleep. Translated to minutes thats 439, 641, and 31. Thus, 439/(641-31) = 72%, not 75%. I realize that's not a huge difference, but for those like myself who have terrible sleep and trying to understand why, accuracy is important. My suspicion is this: One considers the end your last period of sleep as the end of your time in bed, disregarding any additional time you stayed in bed, attempting to sleep, and were either awake or restless. My last period of restlessness was 26 minutes. If you subtract that from the numbers in the equation, it comes out as 75%. And yet I don't know is my suspicion is correct.
