Benford's Law is a mathematical principle that predicts the frequencies for leading digits of numbers within a dataset.
Put simply, Benford's law says that the leading digit in a number sequence is more likely to be a small number like 1, 2, or 3 than a large number like 7, 8, or 9.
Approximately 30% of numbers in a dataset will start with a 1 while less than 5% start with a 9.
The graph below shows this principle applied to a set of company accounts versus the expected Benford distribution. The Sum of Squares calculations compares the fit of each graph. The lower the sum of squares the better the fit to the expected Benford distribution.
The greater a frequency distribution disagrees with Benford's law, the more likely that the number sequence in question is not natural and could suggest potential manipulation.
Benford's law has been used extensively to look for fraud, potential manipulation and unusual numeric patterns in financial records, tax returns, applications, and decision-making documents.