Do as I Say, Not as I Do

I can write about a philosophy a lot more easily than I can live it. Certainly there are people who do as they say they do, but I suspect the majority of those who (like myself) write blog posts on self-help themes don’t practice much of what they preach.

Consider fitness gurus. A lot of physically fit people started with advantages that make staying fit easier in the first place. Research suggests that fitness outcomes are influenced by genetics anywhere from 40-70%, with one twin study finding that genetic factors account for 47-80% of fitness trait variation. My unscientific observations align with this research.

Someone who’s less naturally physically fit can certainly achieve similar (or even higher) levels of fitness, but to do so they would have to override their natural tendencies. This might mean battling their body’s set point weight or fighting against their natural appetite regulation—things that come effortlessly to others.

Fitness is just one version of the same thing: the loudest advocates are often selling people what they want to hear while living by a different set of rules.

Politicians are perhaps the quintessential example of the “do as I say not as I do” mantra. We’ve all seen examples of politicians who champion one cause publicly while privately behaving in contradictory ways. A 2019 study in Political Psychology found that voters are surprisingly willing to forgive this hypocrisy in politicians they support, while harshly judging it in those they oppose.

As a hypothetical example, imagine a politician who campaigns on anti-immigration rhetoric but quietly employs undocumented workers for their household at below minimum wage. They pander to their voter base while simultaneously benefiting from the very situation they publicly condemn.

I try to treat anyone preaching anything—including me—with some suspicion, and pay more attention to what they do than what they say.

If you work in any field related to human behavior, you quickly learn that asking people what they think is often unreliable. This is why political polls can be so inaccurate—they measure what people say, not what they actually do. In behavioral economics, this is known as the value-action gap.

Election outcomes often surprise us because people vote strategically rather than for their preferred candidate, especially in places with winner-takes-all voting systems. Better systems like ranked choice voting can help, though they’re not perfect solutions.

If you want a cleaner read on human behavior, it helps to watch what people do when they don’t know they’re being watched. The observer effect is strong enough that it shapes how we design scientific studies. Double-blind experiments—where neither the subject nor the researcher knows who’s in which group—help control for this bias, but some error is inevitable.

Measuring human behavior is inherently complex because surface-level observations rarely reveal underlying intentions. If you track someone’s food purchases to understand their diet, you might see they frequent McDonald’s but never know if they’re going for the Happy Meal toys rather than the food.

Gurus, preachers, podcasters, whoever: they’re just people. Some of them have useful things to say. I still wouldn’t hand over my judgment to any of them.