How Not to Be Wrong: The Power of Mathematical Thinking

The book opens with the question every math teacher has heard: a student who has spent the weekend on 30 definite integrals asks, «When am I going to use this?» Ellenberg admits that «you'll need it someday» is a lie — but there's a better answer: math is like weight training for an athlete; a footballer never lifts a barbell on the pitch, but that's where the strength comes from. Math, too, works like «X-ray specs», revealing the structure beneath the world's apparent chaos.

At the heart of the introduction is the story of Abraham Wald, a mathematician who fled the Nazis and worked in the secret Statistical Research Group at Columbia. The military shows him the bullet holes on planes returning from Europe: plenty on the fuselage, few on the engines, so the generals want to add armor to the fuselage. Wald asks: where are the bullets that hit the engines? The answer: those planes didn't come back. The armor belongs where the holes aren't. It is the classic case of survivorship bias.

The author shows the same error in mutual funds: if you look only at the returns of funds that still exist, the failed funds that closed drop out of the statistics and the result is inflated — in one study, an apparent 11% annual return falls to 8.9% once the «dead» funds are added back. Ellenberg's central idea: mathematics is an extension of common sense, simply reinforced with sharper tools that reach further.

Chapter one («Less Like Sweden») starts from a Cato Institute blog post: «Why is Obama trying to make America more like Sweden when Swedes are trying to be less like Sweden?» Ellenberg shows that the argument rests on a hidden linear model — «more government is worse» (or the reverse). But the relationship between prosperity and the size of government is a curve: too little is bad, and so is too much. The story of the Laffer curve being sketched on a napkin, Reagan's memory that «after four pictures we all stopped working», and Friedman's tax philosophy bring the point to life.

Chapter two («Straight Locally, Curved Globally») explains when linearity works: Archimedes' method of exhaustion, trapping the circle between polygons with ever more sides; the Pythagoreans' panic over irrational numbers; Newton's insight that every smooth curve looks like a straight line if you zoom in close enough — which the author calls «a one-page calculus lesson». He then moves on to infinity via Zeno's walker who never reaches the ice cream shop and the equation 0.999... = 1.

This section makes clear that the book's strength lies in its rhythm: every abstract idea is tied to a story, a political argument or an everyday example, and formulas appear only when needed. The Turkish translation is readable, though some graphs.

• Each chapter opens with a concrete, often funny, real-life episode
• The author doesn't hide his identity as a mathematician and personalizes the text with childhood memories
• Complex ideas are explained through simple visual analogies (a curve on a napkin, a «tuna sandwich»)
• At times the digressions run long and the main thread can get lost