The book in one sentence: Noise (random variability in human judgment) causes as much damage as bias, and most of us have never thought about it once.
- What Is Noise About?
- What Is the Difference Between Noise and Bias?
- Why Do Experts Disagree So Much About Food and Diet?
- Why Are Your Own Eating Decisions So Inconsistent?
- What Is Decision Hygiene and How Does It Work for Food?
- Is Noise Worth Reading?
- Books Like Noise
What Is Noise About?
Ten nutritionists walk into a room. You give them the same patient file: 38-year-old woman, 30 lbs to lose, no metabolic conditions, sedentary desk job. Ask each one what she should eat, and you get ten different plans. Low-carb. High-protein. Intermittent fasting. Mediterranean. Intuitive eating. Mostly plants. No, mostly animals.
Is the science just unsettled? Sometimes, yes. But Noise offers a different explanation for a lot of that variation: it’s not the science. It’s the judges.
Daniel Kahneman (Nobel laureate, author of Thinking, Fast and Slow), Olivier Sibony, and Cass R. Sunstein (co-author of Nudge) spent years documenting a specific kind of human error that almost no one talks about. Not bias, which pulls everyone in the same wrong direction. Variability: the kind of error that scatters decisions randomly, so that two equally qualified experts examine identical cases and reach completely different conclusions for no good reason. They call it noise, and their central claim is that it causes as much harm as bias while receiving almost none of the attention.
This book is denser and more academic than Thinking, Fast and Slow. It’s written for organizations and researchers as much as general readers. But the core insight transfers directly to the chaos of dietary advice, the confusion of scale readings, and the mystery of why you make completely different food choices on Tuesday than you do on Friday.
What Is the Difference Between Noise and Bias?
Picture three teams of archers shooting at a target.
Team A hits tight clusters near the bull’s-eye. They’re accurate. Team B shoots a tight cluster, but it’s offset to the upper left. They’re consistent, but wrong in a predictable direction. That’s bias. Team C’s arrows scatter across the entire target with no pattern. Some are left, some right, some close, some far. That’s noise.
The math that governs prediction error treats these identically. Mean Squared Error equals Bias squared plus Noise squared. Reduce either one by the same amount and you improve your accuracy by the same amount. They are equally costly. The problem is that we have spent decades studying Team B while barely noticing Team C.
Bias is visible because it has a direction. You can spot it, name it, correct for it. (“This scale reads two pounds heavy, so subtract two.”) Noise is invisible because it has no direction. Different judges are wrong in different ways on different days, and there’s no systematic correction you can make after the fact. The only solution is to build better conditions before the judgment happens.
This distinction matters for anyone drowning in contradictory nutrition information. It’s not all bias. Some of the disagreement you’re seeing isn’t experts being wrong in a particular direction. It’s experts being inconsistent in ways they can’t detect or explain.
Why Do Experts Disagree So Much About Food and Diet?
Noise draws a line between two types of experts that you won’t find in most books about credentials or qualifications.
True experts are people whose judgments can be verified against outcomes. Weather forecasters get rated against actual weather. Surgeons have tracked complication rates. Prediction tournament participants accumulate a verifiable record. Over time, feedback loops separate the accurate from the noisy. True expertise, in this sense, is earned through correction.
Respect-experts are people whose authority comes from credentials, communication skill, and peer esteem, not a verified track record. They construct compelling narratives. They sound confident. But their predictions rarely get checked against outcomes in any systematic way. Without that feedback mechanism, their expertise is largely a performance of confidence.
Most nutrition experts fall into the respect-expert category. A dietitian can practice for 30 years without anyone tracking whether their clients’ outcomes were better or worse than a control group. A doctor can dispense dietary advice with zero outcome verification. A wellness influencer can amass millions of followers without ever being wrong enough for it to matter to their platform.
“Wherever there is judgment, there is noise — and more of it than you think.”
When five credentialed people give you five different answers about whether you should eat breakfast, most of that variation is not five carefully reasoned conclusions from five genuinely different bodies of evidence. Much of it is noise: each expert’s idiosyncratic training environment, their own body’s response to food, their clinical experience, their personality (some are naturally cautious, some are enthusiastic about the latest research), and whatever they read last month. The expert who sounds most certain is not most accurate. They are often just least aware of their own noise.
What to do with this:
- Weight genuine scientific consensus more than individual strong opinions, even from credentialed individuals.
- Look for practitioners who track outcomes (before/after data, client retention, actual lab results over time) over those who communicate confidence.
- Your skepticism about contradictory dietary advice is not a sign of your confusion. It’s a reasonable response to a noisy information environment.
Why Are Your Own Eating Decisions So Inconsistent?
You already know the gap. You eat well for a week and feel in control. Then Friday arrives and you make five decisions in a row you wouldn’t have made on Monday morning. Your knowledge didn’t change. Your values didn’t change. Something shifted that you can’t quite name.
Noise has a name for it: occasion noise, the within-person variability in judgment driven by transient factors the judge cannot detect. The same Israeli parole board that granted parole 65% of the time right after meals granted it near 0% just before meals. Same judges, same cases, same law. Different outcomes because of hunger and fatigue. The same forensic examiner looking at the same fingerprint on different days reaches different conclusions. Same expertise, same evidence, different noise.
Your food decisions are subject to all of this. Hunger, fatigue, stress, time of day, what just happened in the previous hour, how the morning went: all of these shift your judgment without your knowing it’s happening. You are not the same measuring instrument at 9 a.m. Monday and 6 p.m. Friday. Neither is your willpower, your patience, or your ability to care about long-term goals when a short-term one is loudly present.
The standard narrative calls this a willpower failure. You knew what you wanted to do; you didn’t do it; therefore something is wrong with your character or commitment. Noise suggests a more useful frame. Even the most trained, motivated, genuinely trying professionals show enormous occasion noise in their judgments. Trying harder does not fix it. Changing the conditions of the decision does.
The reframe: Eating inconsistency is mostly a noise problem, not a character problem. The question isn’t “why can’t I be more disciplined?” It’s “what conditions produce better decisions, and how do I create more of those conditions more often?”
What Is Decision Hygiene and How Does It Work for Food?
The book’s most practical idea is decision hygiene: a set of practices that reduce noise without requiring you to know which specific errors you’re making. The handwashing analogy is deliberate. When you wash your hands, you don’t know which pathogen you’re avoiding. You don’t need to. The practice reduces an unspecified range of harms across a wide variety of situations.
For food decisions, decision hygiene translates into a handful of concrete strategies.
1. Plan before you’re hungry
Judgments made in advance are less noisy than judgments made in the moment. Deciding what you’re having for dinner at 10 a.m. (before hunger and fatigue have entered the picture) uses a calmer, more deliberate version of your judgment than deciding at 6 p.m. after a long day. This isn’t about willpower. It’s about decision timing.
2. Reduce how many live food decisions you make
Every time you stand in front of the refrigerator without a plan and decide in real time, you’re making a high-noise judgment under conditions that maximize variability. Meal planning, batch cooking, and simple default rules (“on weeknights I always have protein and a vegetable”) all cut the decision load. Fewer live decisions means fewer chances for occasion noise to take the wheel.
3. Separate “what do I want right now” from “what did I decide to eat”
These are different questions with different answers. What you want in a moment of craving is heavily shaped by state (hunger, stress, habit activation). What you can decide to eat is anchored on plans made when you were in a clearer state. Let the clearer state make the decision. Let the tired state follow the plan.
4. Use consistency as a noise-reduction tool
Eat at roughly the same times. Create similar conditions for meals when possible. Consistency reduces occasion noise by narrowing the range of variables that can shift your judgment from day to day. This is why food diaries work partly as a noise-reduction tool: they create a consistent reference point that exists outside the moment.
5. Change the environment before the moment of choice
Don’t keep foods you don’t want to eat regularly in the house. Put the foods you want to eat at eye level and easy to grab. This is decision hygiene in its most physical form. You’re not fighting the in-the-moment decision. You’re eliminating it. The authors call this choice architecture. It’s not willpower. It’s structure.
“Noise, on the other hand, is unpredictable error that we cannot easily see or explain. That is why we so often neglect it — even when it causes grave damage.”
The food diary, meal planning, default protocols, and consistent eating times are all forms of this. None of them require perfect discipline. All of them reduce the surface area for noise.
Is Noise Worth Reading?
Read this if you’ve ever been frustrated by contradictory expert advice, confused by your own inconsistency, or stuck trying harder at something that hasn’t changed. The noise/bias distinction reframes a lot of the struggle around food decisions in a way that is genuinely relieving (it’s not your character) and immediately practical (here’s what to change instead).
Skip it if you want an accessible, conversational read. Noise is academic in structure, heavy on case studies from insurance underwriting and criminal sentencing, and occasionally repetitive about the same examples. The ideas are important; the delivery requires patience.
One caveat: The book is stronger on diagnosis than prescription. The practical tools are real and useful, but implementing structured judgment processes in daily life requires more scaffolding than the book provides. You’ll need to do the translation work yourself. (That translation, for food decisions, is mostly what this review has been doing.)
Books Like Noise
| Book | Author | Best For |
|---|---|---|
| Thinking, Fast and Slow | Daniel Kahneman | The bias half of the error equation — the companion read |
| Nudge | Richard Thaler & Cass Sunstein | Using choice architecture to make better decisions by default |
| Decisive | Chip Heath & Dan Heath | Practical four-step framework for personal decisions with less noise |
| The Art of Thinking Clearly | Rolf Dobelli | The bias catalogue that Noise argues tells only half the story |
| Mindless Eating | Brian Wansink | How environment and cues drive food decisions below conscious awareness |