Apple Health Sucks at Statistics

A quick guide on how not to calculate averages 🧮

Tarek Amr
1 min readOct 29, 2022

Apple pride themselves on their attention to details, and that’s why I had to hold them up to their standards.

I am talking about their Health app in particular, and the way they calculate averages in it.

A yearly log of my weight, tracked in Apple Health app.

This is a chart of my weight over the last 12 months. Each point represents an entry. 🚴‍♂ ️As you can see, after I changed my diet 6 months ago, I started tracking my weight more frequently.

The wrong way to calculate my average weight for the past year is to blindly take the average of all the reported weights. This is a biased average. It puts more weight on the most recent 6 month due to the more frequent entries.

The are multiple ways to deal with that bias. One simple way is to use this curve to find the unreported weights for each day during this year, then calculate the average of these.

This exemplifies why dealing with numbers is not as easy as people think. The problem is that everyone can count, but only few think of what to count

Tarek Amr

I write about what machines can learn from data, what humans can learn from machines, and what businesses can learn from all three.