The SteadyPractice Blog

Personal science, in practice.

Guides on running experiments, understanding your own data, and making decisions based on evidence — not averages.

Start here

·7 min read·Experiment Design

The Confounding Problem: Why Your Experiments Can Fool You

You logged 60 days of data. The intervention looks like it worked. But something else changed at the same time — something you didn't track. Here's how confounding sneaks into personal experiments and what to do about it.

Read
·9 min read

You Can't Blind Yourself: The Placebo Problem in Personal Experiments

In a clinical trial, neither the patient nor the doctor knows who got the real drug. In a self-experiment, you always know. This creates a systematic bias that can make useless interventions look like they're working — and explains why 'I feel better' is not evidence.

Read
·11 min read

The Hidden Variable: Why Most Self-Experiments Give You the Wrong Answer

You started taking magnesium and slept better for a week. Was it the magnesium? Or was it that you also stopped drinking on weekdays, had fewer late meetings, and got a cold that knocked you to bed by 9 PM? Confounders are why most self-experiments mislead you.

Read
·10 min read

Sleep Debt Is a Performance Tax You're Probably Not Measuring

After two weeks of sleeping six hours a night, your cognitive performance is as impaired as someone who has been awake for 24 hours straight. The alarming part: you don't feel that impaired. Your subjective sense of sleepiness adapts. Your performance doesn't.

Read
·11 min read

The Average Patient Doesn't Exist: Why the Same Intervention Works Differently for Everyone

A drug that reduces blood pressure by 10 mmHg on average might lower yours by 20 and your colleague's by zero. The reasons are specific, measurable, and more common than most health advice acknowledges.

Read
·10 min read

The Research Was Wrong: Medical Reversals and the Replication Crisis

Hundreds of standard medical practices have been overturned by later evidence. This isn't a scandal — it's science working. But it means population research is a starting point, not an answer.

Read
·6 min read

Causal Inference for Everyday Life

Did this change actually cause the result? The logic behind causal inference isn't just for statisticians — it's the most useful thinking tool most people never learned.

Read
·10 min read

What p < 0.05 Actually Means (And Why You're Probably Misreading It)

Statistical significance is the most misunderstood concept in health research. It doesn't mean the effect is real, large, or relevant to you. Here's what it actually means — and the three numbers that matter instead.

Read
·9 min read

The Best Books for Self-Experimenters: A Category-by-Category Guide

The books that treat their subject the way J. Kenji López-Alt treats cooking — with controlled tests, honest failure reports, and a refusal to take anything on faith.

Read
·6 min read

How to Design a Personal Experiment That Actually Teaches You Something

Personal experiments don't need to be complicated. But they do need structure. Here's a practical framework for running self-experiments that produce real answers.

Read
·5 min read

Stop Guessing: Why Self-Improvement Needs Personal Experiments

Generic self-improvement advice fails most people not because the advice is wrong, but because it was designed for someone else. Here's a better approach.

Read
·6 min read

The Quantified Self: What the Movement Got Right — and What It Missed

The Quantified Self movement changed how millions of people think about their own data. But 'self-knowledge through numbers' turned out to be harder than it looked. Here's what the movement taught us — and where personal science picks up.

Read
·6 min read

How to Run Your First Sleep Experiment (Without a Lab)

Sleep is the perfect first experiment: it's easy to measure, quick to respond to interventions, and most people have a gut feeling about something that might be affecting theirs. Here's how to test it properly.

Read
·5 min read

The Wearables Trap: Why Tracking Your Data Isn't Enough

Millions of people wear fitness trackers and smartwatches, but most never learn anything actionable from the data. The problem isn't the devices — it's confusing observation with experimentation.

Read
·5 min read

N-of-1 Trials: Why Your Own Data Beats Population Averages

N-of-1 trials compare conditions within a single person over time. They're more informative for personal decisions than any population study — and you can run them yourself.

Read
·4 min read

What Is Personal Science? A Beginner's Guide to Studying Yourself

Personal science applies scientific thinking to your own life — forming hypotheses, running experiments, and drawing conclusions about what actually works for you, not what works on average.

Read

What the research says

Deep dives into the peer-reviewed evidence on each topic.

49 articles

Topic highlights