How Not to Be Wrong: The Power of Mathematical Thinking
Jordan Ellenberg
Why I Recommend This
The bullet holes on returning WWII bombers clustered in the fuselage and wings. The military's instinct was to add armor there. Abraham Wald said: armor the engines—the planes hit there didn't come back.
Ellenberg opens with this story and it sets the tone for everything that follows. Mathematics as a science of not being wrong. Survivorship bias as the paradigm case of how intuition misleads and systematic thinking corrects. A student last semester ran a clean regression on selection-biased data and couldn't explain why the output was wrong. This book addresses the gap between calculation and judgment—demonstrating that mathematical thinking extends common sense, and that mathematical humility (knowing what you can't calculate) matters as much as technical skill.
The Book
Ellenberg argues that mathematics is "the extension of common sense by other means"—a systematic way of not being wrong refined through centuries of collective effort. The book challenges the dominant educational paradigm where mathematics appears as rote memorization, revealing mathematical thinking as an essential tool for navigating a world saturated with data, shaped by algorithms, and characterized by nonlinear relationships.
Through five domains (linearity, inference, expectation, regression, and existence), Ellenberg demonstrates how mathematical thinking reveals hidden structures beneath surface complexity. He covers survivorship bias using WWII bomber armor placement, the replication crisis in psychology, MIT students who beat the Massachusetts lottery through expected value calculations, and Arrow's impossibility theorem proving that perfect voting systems cannot exist. Ellenberg repositions mathematics as civic capability—systematic thinking that augments judgment, available to anyone willing to follow the reasoning.
Passages That Stayed With Me
"Mathematics is the extension of common sense by other means."
The extension runs in both directions—math formalizes intuition, and intuition checks formalism.
"The armour doesn't go where the bullet holes are. It goes where the bullet holes aren't."
The missing bullet holes were on the missing planes.
"Working an integral or performing a linear regression is something a computer can do quite effectively. Understanding whether the result makes sense—or deciding whether the method is the right one to use in the first place—requires a guiding human hand."
The computer runs the integral. You decide whether the integral was the right question.
"One of the most painful parts of teaching mathematics is seeing students damaged by the cult of the genius. The genius cult tells students it's not worth doing mathematics unless you're the best at mathematics, because those special few are the only ones whose contributions matter."
The genius myth narrows who gets to think quantitatively.
"It's not enough that the data be consistent with your theory; they have to be inconsistent with the negation of your theory, the dreaded null hypothesis."
Statistical significance is the beginning of inquiry, not the end.
Read This If...
- • You make decisions based on data but weren't trained in statistics, and want to recognize systematic errors before they compound
- • You suspect that "common sense" often misleads when dealing with probability, correlation, or trend extrapolation
- • You want to understand the replication crisis, p-hacking, and why so many published research findings turn out to be wrong
- • You've been told mathematical thinking is for specialists
- • You need to evaluate policy proposals, business strategies, or media claims that rely on quantitative arguments
Skip This If...
- • You already have formal training in statistics and probability theory—the concepts will be familiar (though the examples might still delight)
- • You want deeper coverage of computational methods, network mathematics, or algorithmic bias—this focuses on foundational reasoning patterns