New Book Club: How Many Monkeys Were in the Room? Reading William Brody's Uncommon Sense

Renaissance oil painting: a dim candlelit hall where rows of monkeys bend over wooden desks, writing; at the front, one monkey in a three-piece suit hands a finished manuscript to a hesitant merchant, while hundreds more keep writing in the shadows

I returned, and saw under the sun, that the race is not to the swift, nor the battle to the strong,
neither yet bread to the wise, nor yet riches to men of understanding, nor yet favour to men of skill;
but time and chance happeneth to them all.
—— Ecclesiastes 9:11 (King James Version, 1611)

The New Book Club: what we’re reading

This is the first post in a series I’m calling the New Book Club. The idea is simple. I pick a book that just came out, read it, share the passages I underlined, and then point the book’s tools back at my own habits.

This round is Uncommon Sense: Rethinking Ordinary Problems in Extraordinary Ways. The author states the problem on the first pages: textbook exercises come with an answer key, and the questions that matter most in life don’t.

The author and the book

William R. Brody’s résumé reads like several people stapled together. He did his bachelor’s and master’s in electrical engineering at MIT, then an MD and a PhD in electrical engineering at Stanford. When the CT scanner appeared in the early 1970s, he moved from cardiac surgery into diagnostic imaging and did early work on CT, MRI and ultrasound. Later he gave up a tenured Stanford professorship to run a medical-device startup, Resonex. In 1996 he became the thirteenth president of Johns Hopkins University and stayed for thirteen years. He also has more than 3,000 hours as a pilot, and he started learning Mandarin in his sixties.

The book grew out of an undergraduate course he taught at Hopkins, also called Uncommon Sense. His co-writer, Mike Field, is a Baltimore writer who wrote for Brody throughout his presidency, from 1996 to 2009.

Johns Hopkins University Press published it in 2026, in twelve chapters (Books.com.tw listing). The copyright page also notes a free open-access edition under a Creative Commons license, so you can read it without paying.

The twelve chapters run from probability through human nature and careers to character. I marked the first three the most. They circle one problem: when someone has a good record, it is hard to tell skill from luck.

64,000 emails, 500 believers

Chapter one opens with a scam. A company called Growth Wealth Trust says it predicts the stock market with AI. On day one it emails 64,000 people. Half the emails say the market goes up tomorrow, half say down. After the close, the half that got the wrong call never hears from the company again. The half that got the right call receives a new prediction, again split half up, half down.

Cut in half, every day. By day eight, 500 people are left, and every prediction they received over seven days was right. Now the company asks for a $500 membership. If only half of them pay, that is $125,000.

Those 500 people did nothing wrong. The record they saw was real: seven for seven. What they never saw were the other 63,500 emails that got it wrong.

Eight side-by-side bars, halved each day from 64,000 people on Day 1 down to a thin bar of just 500 on Day 8; dashed outlines on every bar retain the original height of 64,000 people, and a bracket on the right marks the 63,500 wrong emails no one saw.

How many monkeys were in the room?

Brody puts it in one picture. A monkey in a three-piece suit hands you a wonderful novel and asks you to invest in 10% of its future royalties. Before you pay, he says, ask one question: how many monkeys were typing in that room?

If 100,000 monkeys typed ten hours a day, one good novel tells you the room was big. It tells you nothing yet about this monkey’s talent.

Then he runs the numbers on fund managers. Say 1,000 graduates of one prestigious school go off to manage money, and each has a coin-flip chance of making money in any year. Two to the fifth power is 32, and 1,000 divided by 32 is about 31. After five years, roughly 31 of them have made money every single year, and people call them investing geniuses. The other 969 lost money and moved on. Nobody remembers them.

A grid of 1,000 grey dots arranged 50 by 20, with 31 blue dots scattered among them representing managers who won five years in a row purely by coin flips; below reads 1000 divided by 32 is about 31.

You can take this arithmetic home. When you see a winning streak, divide by the number of people who were guessing at the start. If the answer is still above one, luck alone should have produced at least that many people just like him.

Seven heads in a row is ordinary

Chapter two has a classroom experiment I love. Brody had students draw slips of paper. The “honest” students flipped a coin 200 times and wrote down the results. The “cheaters” made up 200 heads and tails in their heads. He could spot the cheaters at a glance, about 80% of the time.

The giveaway was the streaks. When people invent a sequence, they avoid long runs of the same side, because that doesn’t look random. In 200 real flips, a run of five is close to certain, a run of seven is about 87% likely, and a run of eight is about 69%.

Streaks are normal in randomness. Our brains read them as patterns anyway.

Two rows of 40 coins each: the real-flips row has a streak of 7 consecutive heads boxed in the middle, while the human-invented row alternates frequently with a maximum streak of only 2; three bars below show that in 200 flips, the probabilities of a streak of 5, 7, and 8 of the same side are nearly 100%, 87%, and 69%.

Chapter three goes a step further. Brody shows 90-day price charts for three newly listed companies. One falls to $6. One wanders between $17 and $22. One climbs to $37. You could tell a story about each one. Then he reveals that all three came from 90 coin flips: heads, the price goes up a dollar; tails, it goes down a dollar.

Three side-by-side stock price charts: one slides all the way to $6, one wanders between $17 and $22, and one climbs to $37, labeled below as crash, sideways, and rocket; all three lines were generated by 90 coin flips.

My own ruler got fooled too

The monkey passage reminded me of a mistake I made this September.

I was building a tool to find new uses for indicators I already had. To test whether the tool itself worked, I made a fake indicator on purpose. Random numbers decided which days it fired. It carried no information at all, so it should have scored as “no different from chance.”

The first version of my test gave it a score of 2.4. Statistics commonly uses 2 as the line for “probably not a coincidence,” and it cleared that line.

The problem was what I compared it with. My random control days were drawn from 1994 all the way to 2026, while the fake indicator only fired after 2019. I had measured the difference between two eras. Once I drew the control days from the same stretch of time, the score dropped to 0.7, right where noise belongs.

Two timelines from 1994 to 2026: in the first version, dots for the fake indicator are clustered entirely after 2019 while the control group's dots scatter across thirty-plus years, yielding a score of 2.4; after correction, both rows fall in the same 2019 to 2026 window, dropping the score to 0.7.

It’s the monkey question again. Brody asks how many monkeys were in the room. What I should have asked was whether the monkeys I compared against were typing in the same room.

The same month, a second one caught me from the other side. In early September one of our forecast cards came due. When we wrote it, the card said five sources pointed the same way. When I laid them out side by side afterward, four of the five were reading the same batch of options open-interest data, and the fifth was about margin limits, which answers a different question. There was really one source. The settlement moved 1,350 points the other way.

Five monkeys copying one manuscript are still one monkey. Before I count how many sources agree, I now check how many of them are reading the same page.

Five source circles on the left: four arrows converge into the same 'options open interest' box, while a fifth dashed arrow leads to another 'margin limits' box; the right side notes 5 on the surface, but only 1 truly independent.

So does skill still count?

It’s easy to walk away from these chapters thinking Brody says everything is luck. He doesn’t. He quotes the billionaire John Kluge on the secret of his success: if he said it was all luck, he wouldn’t be honest, “but if I told you it was all strategy, it would be a downright lie.”

He also cites the Harvard sociologist Christopher Jencks, whose 20-year study followed 5,000 families. It found that the two things that shaped income were on-the-job competence and luck, and concluded: “We suspect luck has at least as much effect on income as competence does.”

So what the book asks for is keeping two separate accounts. I copied down one line from chapter one: “One of the hardest things in life can be to recognize in ourselves when we weren’t being skillful and brilliant; we simply got lucky.”

One thing to take with you

The thing that stayed with me: when you see a winning streak, divide by the number of people who were guessing, and only what’s left is skill.

Here’s what I’ve tried. Next time you hear in a group chat, on a show, or from a friend that someone “called it right N times in a row,” take a sheet of paper and write down two numbers: how many times in a row he was right, and your guess at how many people were making calls at the start. Divide the second number by two raised to the first. Write the result in the corner of the page and tuck it into your notebook. If it’s above one, hold off on following him. A month later, pull the page out and check whether he’s still right.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.