
“Statistics is hard, especially when effects are small and variable and measurements are noisy. There are no quick fixes … and a formulaic approach to statistics is a principal cause of the current replication crisis.” — Blakeley McShane

Daily data science tweets from @JohnDCook.

“Statistics is hard, especially when effects are small and variable and measurements are noisy. There are no quick fixes … and a formulaic approach to statistics is a principal cause of the current replication crisis.” — Blakeley McShane

Gauss published a statistical justification for least squares in 1809.

All models are wrong. Some models are useful. -- George Box

Picking the method that has the highest opinion of itself johndcook.com/blog/2011/01/2…

Periodic table of visualization methods visual-literacy.org/periodic…

Vector embeddings of words can approximately satisfy conceptual arithmetic. But when you look at the numbers, this might not seem true. The numbers need to be placed in context. johndcook.com/blog/2026/09/1…

The coefficient of variation is the ratio of standard deviation to the mean.

'It is easy to lie with statistics; it is easier to lie without them.' -- Frederick Mosteller


The word 'statistics' comes from 'state' because first applications were for governments.

Unjustified assumptions of independence are the root of statistical evil.

'BLUE' stands for best linear unbiased estimator, 'best' in the sense of mean squared error.


'The language of science is the language of probability, and not of p-values.' -- Luis Pericchi

An ordinary least-squares regression line passes through the point (mean x, mean y).

A quantile-quantile plot is commonly used as a quick way to check to tell whether data appear to be normally distributed.