Bill Gates IQ: Where the 160 Came From, and What the Maths Actually Says

Search bill gates iq and you will see 160 almost everywhere. Gates has never taken an IQ test. The 160 is not a score he earned — it is his 1973 SAT run through a conversion, with one statistical step left out. Put that step back in and the number drops by about fifteen points. Here is the full working, plus the one piece of evidence that is far better than any test score: a maths problem he solved at 22 that nobody beat for thirty years.

What Is Bill Gates’ IQ?

Nobody knows, and that is the honest starting point. Whether you searched what is the IQ of Bill Gates, what was Bill Gates IQ, or what IQ does Bill Gates have, the answer begins the same way: he has never released a score, no test provider has published one, and he has never claimed a number himself.

So when people ask what’s Bill Gates’ IQ, they are really asking what someone else calculated on his behalf. The Bill Gates IQ number in circulation has an author, and it is not him.

What he did leave behind are two things most public figures never produce. One is a near-perfect score on a hard, standardised exam. The other is an original piece of mathematics that professional researchers took three decades to improve on.

Those two things point in the same direction, and it is not 160.

Where the 160 Actually Came From

Unlike most celebrity IQ numbers, this one is not invented. It has a real method behind it. The method is just incomplete.

In 1973 Gates scored 1590 out of 1600 on the SAT: 790 on verbal, a perfect 800 on maths. That was the pre-1995 version of the test, which was harder to prepare for than today’s and lined up more closely with general reasoning ability.

Now the arithmetic. A 1974 study estimated what the SAT distribution would look like if every American teenager sat it, rather than only the college-bound. On those figures the combined mean was about 770, with a standard deviation near 204.

So the 160 is not made up. Somebody did real arithmetic. They just stopped one step too early.

The Step Everyone Skips

Here is the problem. The SAT and an IQ test are not the same test. They measure overlapping but different things, and they correlate at roughly 0.7 to 0.8 — strong, but nowhere near perfect.

When two measures correlate imperfectly, an extreme score on one does not predict an equally extreme score on the other. This is called regression to the mean, and it is not a technicality. It is the single most common mistake in celebrity IQ estimates.

Think of it in everyday terms. A player who scores five goals in one match is genuinely excellent. But you would not predict five goals next match — you would predict fewer, because part of any exceptional performance is the thing itself and part is everything that happened to line up that day. The same logic applies to converting one test score into another.

The correction is simple: multiply the z-score by the correlation before converting.

So the defensible answer is roughly 142 to 148, depending on which correlation you use. And because the correlation is not perfect, even that carries a wide margin — a 90% band runs from about 129 to 162.

So what is Bill Gates’ actual IQ, as best anyone can say? Around 145, give or take a few points. The IQ level of Bill Gates on that corrected figure is roughly one person in a thousand — a genuinely rare level of ability. It is simply not the one in thirty thousand that 160 would mean.

The Proof He Published at 22

This is the part almost every article on Bill Gates’ IQ leaves out, and it is far better evidence than any conversion.

As a Harvard undergraduate, Gates sat in a combinatorics class taught by Harry Lewis, who mentioned an unsolved puzzle. It goes like this. A waiter is carrying a stack of pancakes, all different sizes, in a jumbled order. Using only a spatula — slide it under any pancake and flip everything above it — how many flips does it take to get the stack sorted, largest at the bottom, in the worst possible case?

It sounds like a party trick. It is a genuine open problem in computer science, and it matters: the same maths is used for routing between processors in parallel computers, and for comparing genomes to work out how species diverged.

Before Gates, the best anyone could guarantee was the obvious approach: bring the biggest unsorted pancake to the top, flip it into place, repeat. That takes about 2n flips for n pancakes.

Gates came back with a better algorithm within days. Working with Christos Papadimitriou, he proved that (5n + 5) / 3 flips always suffice — about 1.67n instead of 2n — and that at least 17n/16 flips are sometimes necessary. The paper, Bounds for sorting by prefix reversal, appeared in Discrete Mathematics in 1979. It is the only academic paper Gates has ever published.

Gates and Papadimitriou also invented a harder variant in the same paper: burnt pancakes, where each one has a burnt side that must finish face down. Their bounds on that version held for about sixteen years.

This matters more than the SAT for a simple reason. A test measures how you perform on problems that already have answers. Solving an open problem measures something else — working out what to do when nobody knows what to do. He was 22, and he did it in his spare time.

Where His Ability Hits a Ceiling

An honest estimate needs limits in both directions, and Gates’ record supplies one.

At Harvard he took the Putnam, the hardest undergraduate mathematics competition in North America. He did not place among the top scorers. Former Microsoft executive Brad Silverberg has also noted that Steve Ballmer outscored Gates on it.

That is useful precisely because it is unflattering. The Putnam selects hard for exactly the kind of raw mathematical ability an IQ estimate is trying to capture. Someone genuinely operating at 160 in that domain would generally show up near the top of that list. Gates did not.

Two data points pushing down, one strong data point pushing up, and they meet in the mid-140s. That convergence is why the corrected range is worth more than either number alone.

What About the Bill Gates IQ 180 Claim?

Some sites list Bill Gates IQ 180 as though it were settled. There is no method behind it anywhere — no conversion, no source, no reasoning shown.

It is worth understanding how extreme that claim is. A score of 145 occurs in about one person in 741. A score of 160 occurs in about one in 31,574. A score of 180 occurs in roughly one person in twenty million.

Nothing in Gates’ record — not the SAT, not the pancake paper, and certainly not the Putnam result — supports a figure that far into the tail. The 180 is what happens when a writer wants a bigger number than the last article they read.

How It Compares to Other Famous Names

Notice the pattern. The popular figure clusters at 160 regardless of who the person is or what evidence exists. That is not four independent measurements agreeing. That is one round number being reused.

Gates is the unusual case here: he is the only one of the four who left behind evidence that can actually be checked. Our guide to famous people’s IQ scores works through why almost every other number on these lists falls apart under the same scrutiny.

Common Misconceptions

Frequently Asked Questions

Related Reading

Last updated: August 2026