Take any large list of numbers. River lengths, stock prices, population counts, the dollar amounts on a company's invoices. Look at the first digit of each value, and count how often each digit from 1 to 9 shows up in front.

You'd expect an even split. Nine possible digits, so each should lead about 11% of the time. But it doesn't work that way. The digit 1 shows up first about 30% of the time. The digit 2, around 18%. The frequencies keep decreasing until you get to 9, which leads less than 5% of the time. This is Benford's Law, and it applies across an unreasonable range of real-world data.

Why 1 Wins

It comes down to how numbers grow. Things in the real world tend to grow by percentages, not fixed steps.

Let’s consider a stock at $1. To climb from $1 to $2 is a 100% gain, and it spends a long time in the "1" range on the way. But once it reaches $9, moving to $10 is only an 11% gain, so it blows through the "9" range quickly. Every quantity that grows this way, money, populations, city sizes, spends more time with a small leading digit than a large one, simply because getting from 1 to 2 is a bigger leap than getting from 8 to 9. Stack up enough of these and the pattern is unavoidable.

Catching Liars With It

Here's where a math curiosity becomes a weapon.

Real financial data follows Benford's Law. Fabricated data usually doesn't, because people inventing numbers spread their digits out too evenly, or cluster them just under approval thresholds, trying to look random. Human intuition about "random" is wrong in a very detectable way.

In the 1990s, an accountant named Mark Nigrini turned this into a forensic tool. Run a set of books through a Benford test, and if the leading digits don't match the expected curve, you have a red flag worth investigating. It's now used by tax authorities and auditors, and it's been admitted as evidence in court. The same test has been pointed at national economic figures and disputed election results to ask, do these numbers look real, or do they look made up?