Does anyone get 90+ sleep score?

[terty]

Curious if anyone gets 90+ or close to 100 in the sleep score? I think the best I got was 87 once when drugged to the moon. Normally 67 to 80 range and can’t help but wonder if that’s accurate, how much better I’d feel getting 20 to 33% better sleep?

[paulo47]

Having had most of the week in the 90s back in May , once the hotter
weather started I consider 80s good !

[ARvdH]

I've had a few sleep scores over 90, including two 100s after totally exhausting myself. It took more than getting a sleep score of 100 to recover so I would not recommend doing this. Getting a score of 80 seems to be optimal for me.

[Tnt2grand]

NINETY????

90????

I would would do ANYTHING TO GET A 75.looks like a avg mid 60s. 

I manipulate the sleep duration since I frequently fall asleep before bedtime while trying to keep up with my favorite MLB Houston Astros while living in NC, so my sleep hygiene/habits are almost impossible. It’s about the only time that I get to spend any time with my wife!

now, having said that, I have Restless Leg Syndrome and just got diagnosed with prostate cancer. Duration has hanged around 6 hrs for as long as I’ve progressed through a series of Fitbit products, at least 10 years. 

My current Fitbit  5 which I’ve replaced at least 4 times (either won’t hold a charge or broken band holding) shows restlessness in the 30% range and awake stages in the 20% range. I’m currently trying to tie steps to these 

I wake up usually once/nie for a trip to the looused to be regularly around 3:30 am then used to could go right. Back to sleep. Since cancer diagnosis, duration took a nosedive to 3-4 hrs but prognosis is good and Prostate cancer is usually very good, most cases curable with 15 yr outlook good barring other health conditions. 

sorry. 

anyway, trying to used everything my Fitbit is telling me to get my duration back up. 

Neurologist has put me on Trazadone to help with anxiety along with the long med list for other conditions which is helping. 

consider yourself very fortunate to get the sleep you do. 

Sleepless. In Carolina 

[Tnt2grand]

I can relate as my post details..

I’m 69 now, bot it was when I was 65 that I finally went to a Urologist. PSA 4.5-5.0.  Treated me after 3 years and I was unhappy with lack of results so last visit I went in with intentions of firing him but he told me he couldn’t help me any longer, had exhausted his resources. 

made an appt with another younger guy up the read and first thing he did was put me on newer meds which helped and did a sonogram for cancer. Viola!!!

schediluled an in house surgery but wanted an Oncologist to weigh in so have delayed recommended urologist’s surgery pending Oncologist opinion. 

Don’t know your whole situation but you’re still young and time is on your side. Many resources in line. Prostrate cancer Research Institute PCRI.com is a great resource if you need info. No one told me until I was 65 that I had prostate pens rot begin with, so you’re ahead of the game!

sleepless in Carolina

[Sleepisgood2000]

I usually get low 90s regularly
And never below 86

I keep the same bedtime every night. I do wear a CPAP so I get really good sleep. I keep my room dark. use the averaging an hour and a half Rem

and and an hour and 20 minutes deep sleep

[Alicia.m]

Me.

Samsung is same. 

[arnie1568]

Great comments. BUT really no good information about the 'HOW & WHY" sleep scores getting into the 90's.
What is the formula?
What are the varopis factors weighed? (Time, Stages, Duration, Quality, etc.)

I consistently get 85 to 89, but in three years, have only seen 90 or above a couple of times. And then there wasn't enough data discrimination to get a handle on what changed, or what I might be able to tweak.

[alexthecat]

@arnie1568 this Help Article explains the factors that influence the Sleep Score. The actual details are proprietary, so there isn't a lot of information available about that. 

[Alicia.m]

To answer your question about the "formula":

I actually consistently scored in the mid-to-high 90s (e.g., 95, 98) on the legacy Fitbit algorithm as well.

My physiological pattern has always been an outlier with a very high deep-sleep ratio,

which is why I get similar results on other devices like Samsung.

The reason we trusted the old Fitbit system wasn't just because of the high scores.

We trusted it because the algorithm was exceptionally stable.

It evaluated your data based on the relationship between your long-term individual baseline and your specific demographic group.

It understood your normal.

However, even though I still receive scores well over 90, I completely distrust Google's new algorithm.

The current system fundamentally ignores the individual baseline. It forcefully applies a flattened, generalized threshold to everyone.

That is exactly why the system is now failing at basic state transitions and creating logical absurdities—like logging active step counts while simultaneously claiming you are in a "sleep" session.

A high score from a broken, baseline-ignoring system is meaningless.

For reference, I am attaching my previous Fitbit data across various sleep durations. As a 43-year-old female, I have physiologically always been the type to extract extremely high sleep efficiency even from very short sleep periods.

Because my sleep architecture is this pronounced, the deviation in Google’s new algorithm remains relatively small for me—just as there was little deviation when cross-validating with Samsung devices.

However, I do not want a meaningless "Sleep Efficiency: 97" score that is calculated purely based on general population averages while my unique individual baseline is completely erased. This fundamental loss of personalized baselining is exactly why Google's new algorithm is so deeply disappointing.

Furthermore, the current Google algorithm suffers from a critical data integrity issue: it actively logs step counts while I am out of bed walking around, yet when I return to sleep 1 hour and 48 minutes later, it retroactively duplicates this exact active period as "awake time during sleep." Logging mutually exclusive data for the same timeline is a fundamental corruption of data integrity.

[SherieB]

Yes, I get 92/93 at least twice a week. 

I'm more puzzled by the introduction of the new app version where the sleep score differs from the old one. It also introduced something called "Sound sleep", which is consistently below the average for me despite all my sleep stages being either in range or above the range.