Analyzing My Weight and Sleep
diet and weight loss | productivity | sleep
Like anyone who has ever been bombarded with magazine headlines in a grocery store checkout line, Kouris Kalligas had a few assumptions about how to reduce his weight and improve his sleep. But data from his wireless scale, diet tracking application, activity tracking devices, and sleep app revealed surprising results.
Excel | Fitbit | Moves | MyFitnessPal | Runkeeper | SleepCycle | Withings Scale
I ran an experiment for three month of data, and the question is what did I do. Out of all this device and tracking which I’ll explain in a moment I spent between 30 to 35 hours analyzing my data and it was in a tool called Excel. It was one year ago, and I also spend around three to five hours actually analyzing my data.
I didn’t track exactly the time of how much I was analyzing my data. But I can tell you it must be three to five hours, and the question is how did I do it and what did I use.
I’m going to show you a bunch of apps and devices I used. It’s not a promotion for them but I think it’s important to say what I used. So I used RunKeeper in terms of tracking my exercise and tracking my cycling. I used myfitnesspal in terms of tracking my nutrition. I used Withings For monitoring my weight and I also used a bunch of trackers for my activity like Fitbit, Fuel band and I also use Move an application which joined Facebook recently. And now I’m developing because of having three and I’m using more on my mobile app so five in general.
The question is out of all the data that I’m tracking and all of these experiences I had, how did it exactly did I look in terms of the data.
So the data in an Excel file you will see it in a moment looks very confusing. I tried to get some correlations out of it. So I tried to see what impacts, what in the end does my activity have to do with my sleep quality. Does my weight driven by a specific thing that I’m not knowledgeable about. This is how that looked, and I’m not going to spend too much time explaining it because it’s impossible. But basically you have the column on the left and on the top and on the left every single biometrics is basically correlations.
Now it’s important to say what did I assume before I did this exercise. Guy talked about what did we do, how did we do it and learn. I think it’s important to say why did you do it and also the assumptions before that.
So these are my assumptions before doing my experiment, and I said okay when my calories go up, my weight go up, when my fat intake goes up my weight goes up. And I had seven simple assumptions and I tried to find out what exactly drives my weight.
You’ll find it actually very surprising that to decrease my weight I actually need to increase my calories. You my say that’s an obvious example in terms of having breakfast in the morning and you cannot burn the calories throughout the day. But actually people normally don’t eat breakfast, and they have this generalization that I want to use calories so I skip breakfast.
I also found out by decreasing my weight I can slightly increase my calorie intake during the day. So basically what this says is the more meals I have during the day the better it is, and this is from a data driven approach.
Again maybe a common sense kind of conclusion but it’s interesting from a data driven approach And also increasing my fat intake during the day I can decrease my weight. So this is a strange result. So if you really take this advice you might start actually eating fat. And also decreasing my weight I can slightly increase my protein intake.
So this is what I found and also interestingly enough I found only from Fitbit, that’s why I mentioned it I can decease my calories I burned. I use Fuel band and I also use Move. It didn’t come out as a strong correlation
Now I’m also very interested in sleep quality and I use sleep cycle and I have no idea what sleep quality definition from sleep cycle is but I like the fact that I can see a trend. So I found basically three conclusions by my sleep quality.
Slightly increasing the calories I burned when I cycle. So the more I exercise, the better I sleep, the higher the sleep quality. And also again an obvious one. And also the more minutes I sleep the better the sleep quality.
Again my second conclusion which might seek obvious to you but it’s interesting that it’s validated from the data. Now the interesting conclusion for the third one is I can increase my sleep quality by slightly increasing my fat mass. Again I’m not saying this is right or wrong because that’s important, but I’m saying that’s what my data tells me. And it’s important to find out what the data tells you because then it’s your choice to actually do something about it and change behaviour based on your knowledge and what you already know.
Now remember the assumptions that I had, the seven assumptions that I had before, about what I found out after I did this exercise, four of the assumptions were actually wrong. So although I had some general assumptions in my mind about calories and weight and fat and weight, the data driven approach actually proved me wrong.
I’m not saying the data approach is correct, but the most important takeaway for me in my tracking experience is that general assumptions don’t work for individuals. And also we should not live our lives on data but with data. It’s important to know if you have a data driven approach and you have results and you take a decision on what the data tells you.
Thank you for your time and you can find me at Kouris Kalligas also at this email. Also tomorrow I’m hosting a breakout session on aggregation platforms and understanding data with Eric Horn. Join us in the morning.