Interactive statistics tool

Interactive Probability Distribution Calculator

This free probability distribution calculator lets you explore and visualize normal, binomial, Poisson, and other statistical distributions in real time. Adjust parameters to see the probability density function (PDF), probability mass function (PMF), cumulative distribution function (CDF), and key statistics update instantly.

Use this tool

Try it now

The workspace opens on Normal Distribution. Every module listed below has its own dedicated guide and the same tool already set to that method.

Interactive tool

The calculator loads as you approach this section so the guide remains fast on mobile connections.

Example preview

Normal lower-tail probability

Inputs
X ~ Normal(μ=500, σ=8); evaluate the lower tail at x=488.
Representative result
P(X<488) ≈ 0.0668, or about 6.7% under the stated model.

Illustrative only. Load the interactive tool to enter your own values and review assumptions.

Choose a supported module

Each module has its own canonical guide, complete server-rendered explanation, and compact tool already set to the method named in the URL.

Normal Distribution →Explore the normal distribution, calculate density and cumulative probability, and learn parameters, assumptions, examples, limits, and common errors.Binomial Distribution →Calculate exact binomial probabilities and learn when fixed trials, constant success probability, and independence make the model appropriate.Poisson Distribution →Calculate Poisson count probabilities and learn how rate, exposure, independence, skewness, assumptions, and exact tails shape interpretation.Exponential Distribution →Explore exponential waiting-time probabilities, rate and scale, memorylessness, constant hazard assumptions, worked examples, and model limits.Gamma Distribution →Explore gamma probabilities with shape and scale parameters, accumulated waiting-time examples, interpretation guidance, and validation limits.Beta Distribution →Explore beta distribution shapes and cumulative probabilities for values on zero to one, with parameter meaning, examples, and cautions.Weibull Distribution →Explore three-parameter Weibull lifetime probabilities, hazard shapes, location thresholds, reliability examples, censoring cautions, and limits.Log-Normal Distribution →Explore log-normal probabilities, log-scale parameters, multiplicative processes, skewed summaries, worked examples, and validation advice.Continuous Uniform Distribution →Calculate probabilities for the continuous uniform distribution and learn how the interval endpoints set its density, mean, variance, and quantiles.Student’s t Distribution →Explore Student’s t distribution, calculate tail probabilities and critical values, and learn how degrees of freedom control its heavy tails.Chi-Square Distribution →Calculate chi-square probabilities and critical values, and learn how degrees of freedom set the shape, right skew, and tail of the distribution.F Distribution →Calculate F distribution probabilities and critical values, and learn how numerator and denominator degrees of freedom control its skewed shape.Bernoulli Distribution →Explore the Bernoulli distribution for a single success or failure trial and learn how one probability sets its mass, mean, variance, and role in bigger models.Geometric Distribution →Calculate geometric probabilities for the number of trials until the first success and learn how memorylessness shapes its long right tail.Negative Binomial Distribution →Calculate negative binomial probabilities for failures before a target success count and learn why it models overdispersed count data.Hypergeometric Distribution →Calculate hypergeometric probabilities for draws without replacement and learn how population size, successes, and sample size shape the count.Cauchy Distribution →Explore the Cauchy distribution whose mean and variance do not exist, with exact quantiles, tail probabilities, and why averaging never stabilizes.Pareto Distribution →Calculate Pareto power-law probabilities and learn how the tail index decides which moments exist and how the 80/20 pattern arises.Laplace Distribution →Explore the double-exponential Laplace distribution with its sharp peak, heavier-than-normal tails, and role in robust and private statistics.Logistic Distribution →Explore the logistic distribution behind logistic regression, its S-shaped CDF, closed-form quantiles, and slightly heavy tails.Rayleigh Distribution →Calculate Rayleigh probabilities for magnitudes of two-component normal noise, from signal envelopes to wind speeds, with exact quantiles.Triangular Distribution →Calculate triangular probabilities from minimum, most likely, and maximum estimates, the standard three-point model for expert judgment.Discrete Uniform Distribution →Calculate probabilities for equally likely integer outcomes, from fair dice to random draws, with exact quantiles and moments.

Distribution relationship map

The families above are not isolated: they connect through exact identities, explicit constructions, and limiting approximations. This static map names the standard relationships and links each one to the pages where it can be explored.

Limits and approximations

These relationships hold in a limit and power the classical approximations; each links to an interactive convergence page where the approach can be watched parameter by parameter.

Binomial → Normal

With p fixed, standardized binomial counts approach a normal distribution as the number of trials grows.

Interactive view →Binomial →

Binomial → Poisson

With the mean np held fixed while n grows and p shrinks, rare-event counts settle onto a Poisson distribution.

Interactive view →Poisson →

Poisson → Normal

As the rate λ grows, the Poisson distribution becomes approximately normal with matching mean and variance.

Interactive view →

Student’s t → Normal

As degrees of freedom grow, the heavy t tails thin out toward the standard normal.

Interactive view →t distribution →

Chi-square → Normal

Centered and scaled chi-square variables lose their skew as degrees of freedom grow.

Interactive view →

Sample means → Normal (CLT)

Averages of independent draws from almost any finite-variance distribution become approximately normal.

Simulator →Normal →

Hypergeometric → Binomial

When the population dwarfs the sample, drawing without replacement is practically drawing with replacement.

Hypergeometric →Binomial →

Special cases

These identities are exact: one family is another family at a particular parameter value, so anything true of the general family specializes immediately.

Bernoulli = Binomial(n = 1)

A single trial is the smallest binomial experiment.

Bernoulli →Binomial →

Gamma(shape = 1, scale = β) = Exponential(rate = 1/β)

The waiting time to the first event is the one-arrival case of the gamma family after matching scale and rate.

Exponential →Gamma →

Chi-square(k) = Gamma(k/2, scale 2)

The chi-square family is a gamma family on a re-labeled parameter grid.

Chi-square →Gamma →

Geometric(trials, p) = Negative Binomial(r = 1, p) + 1

This site’s geometric variable counts the successful trial, while its negative-binomial variable counts failures before success, so their supports differ by one.

Geometric →Negative binomial →

Uniform(0, 1) = Beta(1, 1)

The flat density on the unit interval is the simplest member of the beta family.

Uniform →Beta →

Weibull(shape = 1, scale = λ, location = 0) = Exponential(rate = 1/λ)

A zero-location Weibull with shape one has a constant hazard; matching its scale to the reciprocal exponential rate gives the same distribution.

Weibull →Exponential →

Constructions and transformations

These relationships build one distribution from another by an explicit transformation — the origin stories behind the sampling distributions used in inference.

exp(Normal) = Log-Normal

Exponentiating a normal variable produces the log-normal; multiplicative noise becomes additive on the log scale.

Normal →Log-normal →

Z₁² + … + Z_k² = Chi-square(k)

Summing k squared independent standard normals yields the chi-square distribution with k degrees of freedom.

Chi-square →

t = Z / √(χ²_ν / ν)

A standard normal divided by the root of an independent scaled chi-square gives Student’s t — the shape behind small-sample mean inference.

t distribution →

F = (χ²_{d₁}/d₁) / (χ²_{d₂}/d₂)

The ratio of two independent scaled chi-squares gives the F distribution used by ANOVA and variance tests.

F distribution →

Poisson process: counts, gaps, arrivals

One process at rate λ produces Poisson counts, exponential gaps, and gamma arrival times — exactly, not approximately.

Interactive view →

Gamma-mixed Poisson = Negative Binomial

A Poisson count whose rate is itself gamma-distributed is negative binomial — the standard route to overdispersion.

Negative binomial →Gamma →

Watch the explanation

2:34 min

The player loads only after you press play. You can also watch on YouTube.

Read the complete transcript

How to Choose a Probability Distribution. Choose a probability distribution by ruling out impossible models first. Start with the variable's type and support, then ask what process could generate it. Discrete variables place probability on allowed values. One binary outcome suggests Bernoulli; successes in fixed independent trials suggest Binomial. Counts over exposure may suggest Poisson when the rate is stable. Sampling from a finite population without replacement points instead to Hypergeometric. For continuous variables, support comes before curve shape. Positive waiting times cannot be Normal, and proportions between zero and one need a bounded model. One memoryless wait suggests Exponential; accumulated waits suggest Gamma; lifetime shape may call for Weibull; multiplicative positive outcomes often motivate Log-normal. Use Normal for defensible unrestricted symmetric measurements, not merely because a histogram looks bell-shaped. Use Beta for bounded proportions and Uniform or Triangular for bounded ranges. Inside the explorer, change one parameter at a time. Watch support, center, spread, skew, and tails in both the density or mass view and the cumulative view.

For a continuous model, curve height is density and probability is area over an interval. For a discrete model, each PMF bar is actual probability mass. The CDF answers at most or below. An upper-tail question uses its complement, with endpoint handling checked carefully for discrete variables. For a worked Normal model, set mean five hundred, standard deviation eight, and threshold four hundred eighty-eight. The standardized score is minus one point five. The lower-tail probability is point zero six six eight, so about six point seven percent of modeled values fall below four hundred eighty-eight. Hold the mean and threshold fixed but raise standard deviation to twelve. The same lower tail rises to point one five eight seven because spread changed, not center. A familiar-looking curve is not a fit test. For observed data, check sampling assumptions, diagnostics, tails, parameter uncertainty, and the subject-matter mechanism. Report the family, parameterization, support, probability statement, result, and interpretation. Explore probability distributions free at Distri Scope dot com.

What this probability distribution calculator does

DistriScope’s Distribution Explorer connects formulas to pictures. Choose a supported continuous or discrete distribution, change its parameters, and see the graph update immediately. This is useful when a textbook definition feels abstract or when you need to check how a probability model behaves before using it in an analysis. The workspace keeps parameter controls, plots, summary values, and mathematical notation close together so that each change has a visible consequence.

The explorer is designed for learning and preliminary analysis rather than automated model selection. It helps you ask precise questions: What happens to a normal curve when standard deviation increases? How does the probability mass of a binomial model move when the success probability changes? What cumulative and upper-tail probabilities correspond to a selected value? You remain responsible for choosing a model that fits the data-generating process.

Before you begin

Write down the statistical question, the unit of observation, and the quantity you want to estimate or explain before opening Probability Distribution Explorer. Confirm where the values came from, what units they use, and whether repeated observations are independent. Preserve the original inputs and record every parameter, transformation, and option used in the workspace. This creates a reproducible trail and makes it easier to compare the result with another package.

Treat the graph and numerical output as evidence within a model, not as a substitute for the study design. If a conclusion changes when a plausible parameter or assumption changes, report that sensitivity. Clear documentation is part of statistical accuracy because it allows another person to understand what was calculated, test the same conditions, and identify where an interpretation may need revision. See the DistriScope methodology for formulas, numerical methods, and independent-verification guidance.

Probability distribution calculator capabilities

Parameter-driven visualization

Adjust valid parameters and compare the resulting density or mass function without rewriting formulas or rebuilding a chart.

PDF, PMF, and CDF views

Move between local probability, cumulative probability, and distribution shape so that related statistical ideas stay connected.

Formula and summary context

Review mathematical notation and numerical summaries alongside the plot instead of treating the chart as a standalone illustration.

Point-probability exploration

Use the point calculator to inspect density or mass, cumulative probability, and tail probability at a selected value.

A responsible workflow

  1. 1Select the distribution that matches the type of random variable you want to study.
  2. 2Set parameters within their documented ranges and note the initial center, spread, and support.
  3. 3Switch between density or mass and cumulative views; inspect how the same parameter change appears in both.
  4. 4Use the point calculator to inspect density or mass, cumulative probability, and tail probability, then record the selected value and interpretation in context.
  5. 5Repeat with one parameter changed at a time so the cause of each visual change remains clear.

Worked example: normal spread and tail probability

Start with a normal distribution centered at zero. Hold the mean fixed and increase the standard deviation. The density becomes lower and wider because the same total probability must cover a broader range. Evaluate the cumulative and upper-tail probabilities at a fixed threshold such as x = 1. As standard deviation rises, the CDF at x = 1 moves toward 0.5 and the upper-tail probability increases, even though the mean stays fixed.

Interpretation

The result does not mean the average changed. It means observations are more dispersed around the same average. This distinction matters when comparing measurement precision, process variation, or standardized scores.

How to interpret the result

  • Choose a distribution from subject-matter knowledge before relying on visual resemblance. Count data, waiting times, bounded measurements, and repeated Bernoulli trials imply different supports and assumptions.
  • Use the CDF when the question contains “at most,” “below,” or “up to.” Use tail areas for “at least” questions, and verify whether an endpoint is included for discrete variables.
  • For assessed work, report the distribution name, parameterization, probability statement, numerical result, and a plain-language interpretation. Parameter conventions can differ across software.

Key concepts behind the tool

Density versus probability

For a continuous variable, curve height is a density and can exceed one; probability is the area over an interval. For a discrete variable, the plotted mass at an allowed value is an actual probability. Keeping these meanings separate prevents a common interpretation error.

Parameters and support

A parameter may move the center, change dispersion, alter skew, or control the range of possible values. Support is the set of values the model permits. A model that assigns probability to impossible observations is unsuitable even when its central curve looks plausible.

Quantiles and cumulative probability

The CDF maps a value to the probability of an observation at or below it. A quantile reverses that question: it maps a cumulative probability to a value. Exploring both directions is a reliable way to understand percentiles and probability thresholds.

Limitations and verification

  • Interactive plots are educational aids, not evidence that a distribution fits observed data.
  • Very small tail probabilities can be sensitive to numerical precision and rounding.
  • The tool does not replace study design, sampling assumptions, or domain review.
  • For high-stakes decisions, verify calculations with an independent method and a qualified statistician.

Related DistriScope resources

Compare distribution families →Explore distribution relationships →Calculate statistical tests →

Educational information. Last reviewed 2026-08-09. Calculations should be independently verified for consequential decisions.