Benford's Law reveals that leading digits in real-world datasets follow a logarithmic distribution, not a uniform one. The digit 1 appears as the first digit roughly 30.1% of the time, while 9 appears only 4.6% — a pattern that holds across river lengths, stock prices, physical constants, and even the Fibonacci sequence. The law emerges because natural data spans multiple orders of magnitude, making the logarithmic interval between 1 and 2 larger than between 8 and 9.

The phenomenon was first noticed in 1881 by astronomer Simon Newcomb, who observed worn pages in logarithm tables. Physicist Frank Benford rediscovered it in 1938 at General Electric, validating it across 20,229 data points from 20 categories. Theodore Hill proved it rigorously in 1995, showing that mixtures of random distributions converge to Benford's curve. The formula P(d) = log10(1 + 1/d) defines the unique scale-invariant distribution — multiplying data by any constant leaves the first-digit pattern unchanged.

Forensic accountants exploit this for fraud detection. Humans fabricating numbers avoid leading 1s and cluster around 3, 5, and 7, producing a flatter distribution. The Mean Absolute Deviation (MAD) between observed and expected frequencies flags suspicious datasets. Real accounting records match the Benford curve; invented numbers deviate visibly. The law fails for constrained ranges like human heights or assigned identifiers like ZIP codes.

Benford's Law works because it captures a structural property of multiplicative processes, not a statistical coincidence. Its utility in auditing stems from the gap between human intuition and logarithmic reality — a gap that persists even when fraudsters know the law exists.