7 Unconventional Things I Learned by Tracking My Sleep for 619 Days

At the start of this year I decided that fixing my sleep habits and becoming a better sleeper was the number one thing I could do for my well-being and performance. 

If you’ve ever been interested in digging deeper into your sleep, this email will show you how I did it and, more importantly, how I turned it into something that’s actually useful. 

That’s always the big question.

Is tracking this data actually useful? It’s pretty obvious to most of us when we sleep well, so do we really need an app to tell us that?

In this email I’ll share 7 potentially unconventional things I learned from tracking every sleep metric I could for 619 days, and how I figured those things out. 

Let’s start with the ‘how’...

Every night before I go to bed, I fill out a spreadsheet in Google Sheets. This spreadsheet allows me to enter data for things like the time I go to bed, the time I fall asleep, when I wake up, how I would rate my sleep quality, and more recently has also included my daily Whoop scores. 

(a health and fitness wearables that monitors key physiological metrics 24/7)

I also use this spreadsheet to track my athletic performance along with some metrics that help me gauge my mental and emotional wellbeing. 

All of these data points are recorded in Google Sheets, with a new row for each new day and a new column for each data point. 

If I’m trying to find correlations and other insights, it’s far easier to do that when working with numbers. So where possible, I turn every data point into a number that Sheets can work with easily. 

For example, if I just write 23:30 as the ‘time in bed’, Google Sheets will freak out if I ask it to do maths with that data. So instead I record the time I go to bed as minutes before or after midnight (i.e 23:30 is -30, 1:20 is 80), and the time I wake up as minutes past midnight (i.e 8am is 480). 

The same goes for every other data point I want to integrate into my analysis. I rate my sleep quality as a number from 1-5, my energy level from 1-10, my emotional wellbeing on another number scale, the time I spend exercising as a number of minutes, and so on. 

Once I have two columns with a good amount of data in them, I can use Google Sheets =CORREL function to find the correlation between any two columns. (For example, what is the correlation between sleep duration and energy level? In my case, sleep duration has a 0.15 correlation with energy level. But if I just look at specifically restorative sleep—REM and deep sleep—that data has a 0.27 correlation. So for me, if I want to improve energy levels it’s less about getting more sleep and more about getting more quality sleep.)

I can also use Google Sheets’ ‘standard deviation’ formula (=STDEV.P) to look at the consistency of metrics over time. I’ve always had the goal to make my ‘time up’ have an average standard deviation of 30 minutes, which simply means that my wake-up times are consistent and don't vary by more than 30 minutes from the average each day. 

(using this formula I can see that over the last month the actual standard deviation has been 67 minutes, which tells me I have a lot of work to do in that area.)

Standard deviation and correlation have been the two formulas I feel comfortable with so far, but I know that for someone with better data science skills there’s a LOT more that could be done. 

Still, it’s a start and it gives me some really interesting insights.

The Seven Unconventional Insights

I’ve shared a couple findings already, but here are the seven unconventional insights I found by tracking my sleep. . .

(I’ll start with some more general observations and end with a couple more specific insights based off of my data.)

1. Tracking your sleep is useful because it makes you accountable to ‘future you’

There’s something pretty painful about looking back at my sleep data and realising how many days I did a horrible job of prioritising healthy sleep habits. 

On days where I’m tempted to sleep in longer or go to bed later, it helps me to remember that any decision I make will end up as part of this data set and even potentially shared publicly (like in this email.) 

I want my data to tell a story about me that aligns with the person I think I am. If I think I’m a disciplined person, but my data doesn’t show that, it creates the desire for me to continue working at it until it does. 

2. Tracking your sleep is a daily reminder to prioritise what’s most important to you

Even if the data doesn’t give you any crazy insights, the fact that you’re tracking it reminds you every single day that it’s something important to you. 

3. The ‘days in between’ metric can help you see the bigger picture

This metric simply tells me the average number of days in between days where I accomplished my healthy sleep habits. 

For example, when I first started trying to remember to turn my phone off an hour before bed, I would rarely do it. Maybe once every few weeks. In other words, the ‘days in between’ me accomplishing this habit was probably at least 14 on average. 

But over time, and because I get a daily reminder to focus on this habit when I fill out the sheet and log whether I did it or not, I’ve gotten slightly better at remembering to do this. 

Now on average it’s probably once every three or four days that I turn my phone off an hour before bed, so my  ‘days in between’ metric might read 3.5. 

Although I’ll still feel annoyed on days where I don’t achieve my goal, it helps to see that day as part of the bigger picture. 

If I can see that the trend is going in the right direction—down from 14 to 3.5, and hopefully down again to 1 or less soon—I don’t need to get so caught up in the discouragement of the moment.  

Focusing on this macro trend is more encouraging than getting lost in the day to day. 

4. It took longer than I expected, but my sleep is improving

I guess I thought that tracking my sleep would change me overnight. In reality, it’s taken many months just to get somewhat better at it. Last year I would go months without going to bed before midnight, but now it’s rare to go a week without doing so. I’m happy with the trajectory though and I think it would be much worse without taking the steps I did. 

I think most people would probably do better than me. I don’t think I’m very good at sleeping well. I think it’s just the fact that if you track something and think about it for 619 days, it’s very hard not to get at least a little bit better at it. 

5. Sleep affects my energy, energy affects my mood

When I was trying to find correlations between lifestyle factors and my emotional states, the one that stood out the most was my energy level. 

My self-reported energy level has a 0.45 correlation with my reported feeling of emotional well-being that day, which is a way stronger connection than most everything else I could see. 

What’s fascinating is the indirect nature of some relationships. For example, sleep duration itself doesn’t show a strong direct correlation with emotional well-being. However, it does have a noticeable link with my energy levels. This suggests that by improving my sleep, I can boost my energy levels, which in turn seems to contribute to better emotional well-being.

It’s a reminder that not all relationships are straightforward, and it’s useful to really try to understand the data before jumping to conclusions. 

6. Sleep affects performance, and correlation isn’t causation

I mentioned this a few weeks ago when analysing my marathon, but it’s worth repeating here…

The time I fall asleep has a -0.26 correlation with the strength of my runs, suggesting that later bedtimes might be linked to feeling less strong during runs. Similarly, the time I wake up has a -0.21 correlation, hinting that earlier wake-up times might lead to stronger runs.

One surprising observation is the weak negative correlation between the amount of sleep and running strength (-0.13). Even the amount of restorative sleep (REM and deep sleep) shows a weak negative correlation (-0.14) with running strength. At first glance, this seems counterintuitive, as we expect better sleep to enhance physical performance.

However, this might reflect the limitations of correlation analysis. For example, on days when my body is more fatigued, I might sleep longer, which could skew the data—making it look like more sleep is associated with weaker runs.

Interestingly, my self-reported energy level has a 0.28 correlation with the strength of my runs. This positive relationship makes sense: when I feel more energized, my body performs better. Since my sleep duration does correlate positively with energy levels, it’s reasonable to conclude that better sleep indirectly supports stronger runs by improving my energy. The direct link may be obscured by other factors, such as the need for recovery sleep on more fatigued days.

7. Sleep data isn’t everything, and there are things to be cautious about

One big critique with wearables is that some people might find that viewing their sleep scores when they wake up can be stressful. This can even lead to ‘illusory fatigue,’ where the body feels more fatigued due to psychological factors—like believing it should be tired after seeing a low score—even if it’s not physiologically true.

If this is the case, it’s important to find ways around this—like reviewing your automated sleep scores weekly rather than daily. (although I would still recommend manually logging activities daily if you choose to do that part, you’ll probably forget a lot of the details by the end of the week)

Secondly, people who say “It’s pretty obvious to most of us when we sleep well, so do we really need an app to tell us that?” are also right in a way. 

I think all of us can just tell, whether we’re tracking it or not, that sleeping well feels good. 

The point of tracking all this data isn’t to replace our natural ability to pay attention to our body and make changes based on our own feelings. 

It’s almost certainly better to sleep well and not track it than sleep badly but be able to dissect it in a hundred different ways. 

All the data you generate should be used as just another tool to help you sleep and live better. To the extent that it doesn’t do that, don’t stress it.

But if you’re willing and have the capacity, I’ve found that doing this has been one of the best decisions I’ve ever made for my sleep—and as a result, so many other aspects of my life. 

Until next week, 

Benji and Jacob

This week’s content sponsored by 1440 Media. . .

We personally find being overwhelmed with news in our inbox stressful. But there definitely are good reasons for staying informed, and this company seems pretty solid. So if you’re into this kind of thing, check it out…

Seeking impartial news? Meet 1440.

Every day, 3.5 million readers turn to 1440 for their factual news. We sift through 100+ sources to bring you a complete summary of politics, global events, business, and culture, all in a brief 5-minute email. Enjoy an impartial news experience.

Join for free today!

If you enjoyed today’s newsletter, feel free to share it with a friend!

If someone shared this with you, go to whatcounts.io to subscribe and receive weekly emails like this.