73 free tools — instant, accurate, and no sign-up required.
Mean, median, mode, standard deviation, variance, IQR, skewness, kurtosis, and more.
PDF, CDF, z-scores, and tail probabilities for any normal distribution.
Convert a value to a z-score, find its percentile, and tail probabilities.
One-sample, two-sample (independent), and paired t-tests with interpretation.
Goodness of fit and independence tests, with Cramér's V effect size.
Compare means across three or more groups. F-statistic and p-value.
Non-parametric alternative to the independent t-test for skewed or ordinal data.
Check whether data are normal, with a Q-Q plot, histogram, skewness and three tests.
Non-parametric alternative to the paired t-test, with exact p-values and an effect size.
Non-parametric alternative to one-way ANOVA, with Dunn's post-hoc comparisons.
Exact test for small 2×2 tables, with the odds ratio and its confidence interval.
Compare two proportions or conversion rates — ideal for A/B tests.
Bonferroni, Holm and Benjamini–Hochberg adjusted p-values for many tests.
Peek at your A/B test any time: always-valid p-values and intervals.
CI for a population mean (t) or proportion (z), at any confidence level.
Required n for estimating a mean or proportion at a target margin of error.
How many visitors each variant needs from your baseline rate and target lift.
Responses needed for a chosen margin of error and confidence level.
Participants per group to detect a difference between two means (t-test).
Subjects needed for a before/after (paired) or one-sample t-test.
Cohen's d and Hedges' g for two-sample, one-sample, and paired designs.
Pearson and Spearman coefficients with p-value, 95% CI, and scatter plot.
Slope, intercept, R², residuals, and a fitted regression line chart.
Exact and cumulative binomial probabilities, with distribution table and chart.
Probability of k events given an average rate λ, with table and PMF chart.
Two events — AND, OR, conditional, and "at least one" — with a live Venn diagram.
Compute nPr, nCr, and factorials exactly, with a binomial-coefficient chart.
Drag sliders and watch the normal, binomial, Poisson, and t distributions change live.
Wilson error bars for a model's accuracy score, plus items needed for a target margin.
Is model A really better than B on the same eval? Exact paired test + CI for the gap.
How many eval items you need to detect an accuracy gap between two models.
Unbiased pass@k for code and reasoning evals (Codex-paper estimator).
Agreement between two raters — e.g. an LLM judge vs humans — with weighted kappa.
Agreement among 3+ raters or LLM judges, with a bootstrap CI.
Reliability with missing ratings and nominal, ordinal or interval scales.
Bradley-Terry leaderboard with 95% CIs from head-to-head wins.
Bootstrap error bars for F1, precision, recall and accuracy.
Compare two models on F1, macro-F1 or accuracy with a CI for the gap.
Does confidence match accuracy? ECE, Brier score and a reliability diagram.
Error bars for average judge scores, plus a paired comparison of two systems.
Drag sliders and watch the normal, binomial, Poisson, and t distributions change live.
Pick any population, draw samples, and watch the sample mean become a bell curve.
See Cohen's d as two overlapping groups, with U3, overlap, probability of superiority and NNT.
A multi-armed bandit game: find the best slot machine in 100 pulls and out-earn a Thompson sampling algorithm. Daily challenge.
A stats memory game: a bell curve flashes, you recreate its mean and SD. Daily challenge and 2-player mode.
A stats game: see a scatterplot, guess r, get scored. Build real intuition for correlation.
Turn t-test, ANOVA, chi-square, correlation and nonparametric results into a correctly formatted APA 7 sentence.
Exact p-values from z, t, chi-square, F or r, with an APA-formatted result.
Find the sample size you need or the power of your study, with a power curve.
Q1, Q3 and the IQR by the Excel, R, SPSS and TI-84 methods, with a box plot.
IQR fences, z-scores, modified z-scores and Grubbs test side by side.
Shape statistics with standard errors. Matches Excel and SPSS.
What a positive result really means, from the base rate, sensitivity and specificity.
Check whether a comparison reverses when you split the data into subgroups.
How much change in extreme scores is expected by chance alone.
How the chance of a false positive grows with the number of tests.
Run a Welch or Student t-test from summary statistics, with no raw data.
One-sample and two-sample z-tests with a confidence interval.
Main effects, the interaction, partial eta squared and an interaction plot.
Nonparametric repeated measures, with Kendall's W and post-hoc tests.
An exact test for paired data, using only the direction of change.
Exact t, z, chi-square and F cut-offs for any alpha and df.
The plus-or-minus figure for a survey percentage or a mean.
Wilson, exact, Agresti-Coull and Wald intervals side by side.
From raw data, an SD and sample size, or a proportion.
Compare variability across different units.
The value at any percentile, and the percentile rank of any score.
E(X), variance and SD from outcomes and probabilities, with steps.
The 68-95-99.7 ranges, or check your own data against the rule.
Answer a few questions and get the right statistical test with a link to its calculator.
Looking for critical values? Browse the statistical tables →
New to statistics? Start with the plain-English guides →
Revising? Grab the statistics formulas cheat sheet →
Teacher or blogger? Embed any calculator free on your own site →
Not sure which test you need? Try the interactive test chooser →