Why "Quantified Self" Fails
photo credit: Quantified Self #10 by Sebastiaan ter Burg, additional edits mine.
Quantified self (QS) is a loose movement which believes in closely tracking data about yourself in order to make more informed decisions about how to guide your future actions.
We can assume that QS has existed in numerous incarnations throughout history - even something as simple as writing a journal or having a mindfulness habit technically qualifies as quantified self.
In the more modern sense, quantified self refers to the recent batch of consumer gadgets designed to make it easy to gather data about, well, yourself. This includes simpler devices like heart rate monitors and step counters, but also more complicated devices like FitBits, Apple Watches, and more.
Naturally, quantified self has been very closely associated with the fitness industry. Health and fitness enthusiasts have always looked for ways to improve their performance and health, and many of us use detailed tracking methods already, including logs of our diet, weight, body composition, and workouts. It’s not too much of a step from there to start gathering other data on our bodies.
When QS was being defined as a movement, there was great enthusiasm that it would solve many of the problems that exercisers have. Initially, it seems promising.
In Thinking In Systems, Donella Meadows tells a story of a scenario in Dutch houses in Amsterdam in the 70’s:
“During the oil embargo and energy crisis of the early 1970s, the Dutch began to pay close attention to their energy use. It was discovered that some of the houses in this subdivision used one-third less electricity than the other houses. No one could explain this. All houses were charged the same price for electricity, all contained similar families.
The difference, it turned out, was in the position of the electric meter. The families with high electricity use were the ones with the meter in the basement, where people rarely saw it. The ones with low use had the meter in the front hall where people passed, the little wheel turning around, adding up the monthly electricity bill many times a day.”
We can imagine QS should work in the same way: by providing us active feedback and data about our activity, we should be encouraged to do more of it.
Indeed, QS does accomplish this for certain kinds of people. By providing themselves with more data, dedicated QS’ers can do more with their efforts at self measurement.
Quantified Self For General Health
That being said, I don’t think QS is any kind of magic bullet, and I think that in the years since its heyday, the allure has faded quite a bit. For example, Dick Talens’ Why “Quantified Self” Is Bullshit ruffled quite a few feathers when it first came out, but also made a few serious points.
One of the big promises of QS is that if you have more data, you can make better decisions. However, having more data doesn’t always mean better decisions - just because you know the right course of action doesn’t always inspire you to take it. I might know that I’m not supposed to eat a slice of cake sitting in front of me, but that doesn’t mean I won’t eat it.
One of the ways that fitness isn't like the example of the Dutch houses is that fitness isn't as simple as flicking a light switch on and off, or remembering to turn your television off right away when you're not using it. It requires a lot more energy and a greater change in your habits to take yourself to the gym, or to make the conscious effort to go for a walk. Not to mention, it involves a much greater investment of your time! In such cases, a simple QS device probably won't provide enough of a motivator to make bigger changes in your habits.
Likewise, having more data on your steps taken, metabolism, and heart rate won’t necessarily inspire you to move more or instantly give you other related knowledge you need to succeed in fitness. Just because I have more data doesn’t mean my performance will instantly improve.
As Dick puts in the above article: fitness is a human problem, and not a tech or systems problem. What that means is that it’s much more complex than simply gathering data. Motivation, lifestyle management, and accountability are equally important if the goal is to actually execute on your plan to get in shape.
Adding a FitBit, Apple watch, or other QS device into your life isn’t going to immediately make positive improvements.
Quantified Self For Performance
Another common argument for QS is that it isn’t actually intended to be a cure-all for the problems of people who aren’t terribly motivated in the first place. Instead, the intention is to allow dedicated QS’ers and serious fitness enthusiasts the ability to make more informed decisions - allowing them to further advance their own results.
This argument is essentially true - while beginners won’t find raw data particularly motivating, people who enjoy the data and the activities it reflects will be able to use it to regulate their activity for maximum effect.
At the same time, I want to caution against thinking that data can solve all our performance problems either.
I once read an email from a young man who very compulsively tracked data about himself. He attached for me the spreadsheet in which he recorded it all.
This spreadsheet contained all imaginable data - every single meal he’d eaten for months, with macro breakdowns and calorie counts for all of it, every single exercise he’d done in every single workout, graphs of his progress in all kinds of variables, detailed tracking of his bodyweight and body composition, and of course data from his FitBit as well. This spreadsheet was massive, and I can only imagine how much time he spent creating and maintaining it.
What had he gotten out of the spreadsheet? Was he a super strong lifter? A world class athlete? A super genius?
Ultimately, he got nothing out of it.
In fact, the reason he had written the email was that his friends had gently implied to him that he should stop wasting his time gathering this data - and he was coming to ask if there was some way that someone else could put this data to use, making him a subject in a scientific case study or something similar.
He had more data about himself than many do, but this didn’t immediately make him a better or stronger lifter. He was actually a pretty average lifter - stronger and more jacked than the average person, but certainly no Arnold Schwarzenegger or Bruce Lee. Knowing the underlying data doesn’t immediately change it, particularly if you’re measuring data that isn’t particularly relevant.
Quantified Self As Bad Science
One problem with gathering so much data is that it’s an example of poor science.
When reading a study, you typically see statistics labelled as confidence intervals, or CI, as well as p-values. You don’t have to understand how either of these work - they’re complex statistical concepts that even scientists get wrong. People go to school specifically to understand these, so it would be silly for me to expect you to understand them completely.
These statistics are used to essentially give a sort of rating to the data in the study, telling us how likely it is that the data gathered is an accurate representation of the underlying phenomena being studied. When we create a study, poorly designed studies have a high chance of creating false positives - data that seem to say one thing when the reality is quite different.
In scientific testing, we typically limit the number of variables being studied in a single study. If you study too many data points at once, you raise the risk of creating false positives, which make it easy to find imaginary trends and draw false conclusions.
Here’s an example from John Bohannon’s experiment on fooling the world into thinking chocolate is good for us:
“Here’s a dirty little science secret: If you measure a large number of things about a small number of people, you are almost guaranteed to get a “statistically significant” result. Our study included 18 different measurements—weight, cholesterol, sodium, blood protein levels, sleep quality, well-being, etc.—from 15 people... That study design is a recipe for false positives.
Think of the measurements as lottery tickets. Each one has a small chance of paying off in the form of a “significant” result that we can spin a story around and sell to the media. The more tickets you buy, the more likely you are to win. We didn’t know exactly what would pan out—the headline could have been that chocolate improves sleep or lowers blood pressure—but we knew our chances of getting at least one “statistically significant” result were pretty good.”
The solution is not to measure 18 different variables until you find something that fits - it’s to monitor just a handful of very important variables that are accurate measures of progress.
QS suffers from the same problem - if you’re gathering a ton of data, it’s easy to start finding false results in your own progress. Worse, since you’re a single data point (n=1) your own data doesn’t necessarily have a ton of statistical power - meaning it’s very hard to generalize your experiences to anyone else. Potentially useful if you’re a serious athlete, but you're still probably just creating a lot of white noise.
This means QS is a huge red herring. While gathering lots of personal data and using it to guide your behavior and decision making is a noble ideal, I’m just not sure it delivers. Existing technology is prone to error and isn’t really capable of measuring very many variables to begin with. A lot of that data, like in my example of the person who sent me his spreadsheet above, doesn’t necessarily have any practical usage.
I think that if you have specific need for a particular QS device - for example, you’re a runner and want a nice heart rate monitor to help determine heart rate training zones during your workout - then it can be a clear net positive.
However, if the specific data point is less relevant to you (a heart rate monitor to a weightlifter) or if you’re just a beginner who struggles more with other issues like motivation and accountability, QS devices will probably not have too much of a benefit.
That said, I always recommend the softest version of QS - mindfulness, journaling, and basic diet/exercise tracking - since this tends to be the most useful in terms of providing general feedback that you can use to structure your decisions. In these cases, no fancy devices are needed.
Enjoyed this post? Share it on Facebook or Twitter.
Takeaway Points:
Quantified Self is a movement which focuses on the use of technological devices to gather data about our bodies in order to make more informed decisions about our health and fitness.
However, knowing data doesn't necessarily immediately change it. A QS device alone may not be sufficient to inspire serious changes in habits for beginners. Ultimately, fitness is a human problem, and not a data problem.
Serious exercisers may get more out of it. But at the same time, they may also fall prey to wasting their time focusing on unimportant white noise.
It's generally better to focus on a handful of important variables than to focus on a lot of unimportant ones, which is likely to create false positives and lead us to false conclusions.
QS is a bit of a red herring - I highly recommend tracking in a basic sense, but it's generally not necessary to track extensively the kind of variables that QS allows us to examine.
Further Reading:
Are you interested in perfecting your deadlift and building legendary strength and muscle? Check out my free ebook, Deadlift Every Day.
Interested in coaching? Inquire here. You can also subscribe to my mailing list, which gets you the free GAINS exercise program for maximizing strength, size, and endurance.
