Key Takeaways

  • Fund manager Joel Greenblatt conducted an experiment where a room of people guessed the number of jelly beans in a jar.
  • When participants wrote their guesses privately on paper, the group average was 1,771 beans, just five away from the true count of 1,776.
  • When participants stated their guesses out loud, group anchoring caused the collective average to miss the mark by 50%.
  • Market noise, analyst price targets, and social consensus consistently pull smart people away from independent, accurate calculations.

The Jelly Bean Experiment: Private Math vs. Public Noise

Shaan Puri recounted a classic demonstration run by legendary value investor Joel Greenblatt. Greenblatt walked into a crowded auditorium, placed a large jar filled with jelly beans on the stage, and asked everyone in the room to estimate how many beans were inside.

In the first version of the test, every person worked in total silence. They looked at the jar, did their own visual math, wrote down a number on a slip of paper, and handed it in. Greenblatt collected the slips and calculated the mathematical average of the room.

“That's the average of all the guesses,” Puri explained. “Now, guess what? That's only five off from the actual. There's 1,776 beans in there.”

Independently, a crowd of regular people produced an estimate of 1,771 beans. The wisdom of crowds worked because individual errors cancelled each other out. Some guessed too high, some guessed too low, but their independent observations centered squarely on objective reality.

Then Greenblatt altered one variable. He asked a fresh group of students to look at the exact same jar, but this time, each person had to stand up and shout their estimate out loud before the next person spoke.

Chaos followed. The first student made a wild guess. The second person anchored to that number. The third adjusted slightly up, and the fourth panicked and adjusted down. Instead of consulting the jar, people began reacting to the voices around them.

“When they all said their guesses out loud, they went from being off by five to being 50% off,” Puri noted. Social proof corrupted the data feed. The moment individuals heard other opinions, their brains stopped calculating volume and started managing social conformity.

Why the Market Destroys Independent Thinkers

This dynamic explains why so many smart operators lose money when they touch public equities, crypto, or late-stage venture rounds. They stop looking at the business and start tracking the room.

“This is the stock market,” Puri said. “For most of you, if you were able to independently think and analyze a business and try to come up with what you think the true value of that business is, you're smart people, you do pretty well.”

In public markets, the shouting never stops. You open your phone and see quarterly earnings hot takes, Wall Street price target upgrades, and panic posts on Twitter. Each headline acts like a student yelling an absurd estimate across the auditorium. When you see a stock surge 40% in a month, your brain anchors on the momentum rather than cash flow or revenue quality.

Greenblatt's core investing lesson is simple: isolate yourself from the noise before you make a decision. As Puri put it: “Look at the jar and write down the number on the piece of paper and know that on average you will be much closer to the true value than if you just listen to what everybody is saying.”

What to Do With This

Before you review an investment, a competitor's valuation, or a major product decision this week, write your thesis and target valuation in a blank document before checking secondary opinions. Do not open Twitter, do not check analyst notes, and do not ask your peer group for their gut check until your personal numbers are locked on paper.