A crowd of villagers guessing the weight of an ox in 1906 landed closer to the truth than most individual experts could manage on their own.
Statistics today feels like a modern science, full of computers and complex models. But some of its earliest and most convincing demonstrations came from remarkably simple, almost accidental experiments.
Born from the needs of growing states
Modern statistics really emerged in the 18th century, driven by the practical needs of industrializing states that needed reliable numbers to govern effectively. The word statistics itself traces back through the Italian statista to the medieval Latin statisticum collegium. Early statistics dealt almost entirely with demographic and economic counts, tools built specifically to help run a country rather than to explore pure mathematics.
The mathematical theory underneath statistics did not develop seriously until the 19th century, and its formal connection to probability theory took just as long to solidify. Three major waves of development, spanning roughly the turn of the 20th century through the 1930s, shaped statistics into something recognizable today.
Counting bricks before a siege, and counting the dead in London
Long before any of that formal structure existed, the Greek historian Thucydides used a statistics style method during the siege of Plataea in the 5th century BCE. Athenian soldiers repeatedly counted the exposed bricks on a section of the enemy’s unplastered wall, then used the most frequently repeated count to estimate the height needed for their scaling ladders.
Centuries later, in 1662, John Graunt and William Petty built the first genuine mortality tables and population counting methods. Graunt studied London’s death records, found roughly 13,000 deaths per year, and used that ratio to estimate the city’s total population at around 384,000, an impressively accurate early statistical achievement.
Founders who gave statistics its modern shape
Francis Galton is widely regarded as the principal founder of modern statistical theory, introducing the core concepts of standard deviation, correlation and regression. Karl Pearson built on that foundation, establishing the world’s first university statistics department at University College London in 1911.
Ronald Fisher led statistics through its second major wave, publishing Statistical Methods for Research Workers in 1925 and The Design of Experiments in 1935. Pierre Simon Laplace, working much earlier, developed Bayesian statistics and applied Bayes’ theorem to both celestial mechanics and medical statistics of his era.
The lady, the tea, and the ox
Fisher illustrated the core principles of experimental design with a now famous story about a woman who claimed she could tell, simply by taste, whether milk or tea had been poured into her cup first. He designed a rigorous test specifically to check her claim, a story that still opens discussions of experimental design today.
Galton’s own contribution came from an unlikely source: a country fair. He examined 787 separate guesses of an ox’s weight submitted by fairgoers in 1906. The median of every guess combined came out to 1,198 pounds, against an actual weight of 1,208 pounds, a striking early demonstration of what is now called the wisdom of crowds.
From counting bricks under siege to guessing an ox’s weight at a country fair, statistics has always drawn its power from a simple idea: individually shaky guesses, gathered in large enough numbers, can add up to something remarkably close to the truth.

