Discrete distribution generator

How it works

1. Choose a distribution. Bernoulli returns 0 or 1; binomial counts successes in a fixed number of trials; Poisson models nonnegative event counts.
2. Set probability between 0 and 1, a binomial trial count up to 200, or a Poisson rate from 0.01 to 50. Weighted outcomes use label | weight lines.
3. Generate a sample, inspect the observations and export the values with their recipe.

Parameters mean different things

Bernoulli has expected value p. A binomial model with n independent trials has expected value n × p and variance n × p × (1 − p). A Poisson model has mean and variance λ. RandJoy sums Bernoulli trials for binomial samples and uses a product-of-uniforms method for Poisson samples within the supported rate range. Weighted labels are sampled in proportion to their positive weights.

Your inputs stay with you

Your inputs and generated results stay in this browser. Settings and five recent runs use local storage; clearing browser data removes them. Export a copy to keep it. Hosting may record standard page-request logs, but generated rows and participant names are not sent to the server.

Can I use this to predict real events?

A distribution describes assumptions, not evidence that a real process follows them. For example, a Poisson model assumes a constant rate and independent arrivals. Use the output to explore a model or test code, and check real-world assumptions separately.