← Math & statistics

Probability Simulator — coin, dice, roulette (theory vs empirical)

Fix the RNG seed and run coin, weighted dice, or roulette experiments. Watch theory vs empirical results, confidence intervals, convergence bands, and step logs update in real time.

Designed for probability lessons. Deterministic seeds keep classes in sync, and CSV/LaTeX exports make documentation easy.

Other languages 日本語 | English | 简体中文 | 繁體中文 | 繁體中文(香港) | Español | Español (LatAm) | Español (México) | Português (Brasil) | Português (Portugal) | Bahasa Indonesia | Tiếng Việt | 한국어 | Français | Deutsch | Svenska | Suomi | Dansk | Norsk bokmål | Italiano | Русский | हिन्दी | العربية | বাংলা | اردو | Türkçe | ไทย | Polski | Filipino | Bahasa Melayu | فارسی | Nederlands | Українська | עברית | Čeština

Choose coin flips with custom p(Heads), weighted six-sided dice, or European/American roulette bets. The simulator compares theory and empirical ratios, shows Wilson 95% intervals for binary outcomes, tracks convergence, and records each calculation step.

For handouts, export raw counts as CSV or copy the LaTeX summary. Use the shareable URL to reopen the same scenario later.

Controls

General settings
Classroom presets
Coin settings
Embed this calculator

Set the conditions, then run the simulation.

The fixed-seed LCG is provided for reproducible teaching. It is not suitable for cryptographic or security-sensitive randomness.

Roulette is included for probability education and does not improve gambling odds.

Result

Set the conditions, then run the simulation.

How it's calculated

    Distribution (theory vs empirical)

    Bar chart comparing theoretical probabilities with observed proportions.

    Running estimate vs trials

    The shaded area shows a pointwise Wilson 95% interval at each trial count; it is not a simultaneous confidence band.

    Bar chart comparing theoretical probabilities with observed proportions.

    Probability simulator chart data table
    OutcomeTheoryExpected countObserved countEmpiricalDifference (percentage points)

    Line chart showing the running estimate and pointwise Wilson interval by trial count. The shaded area shows a pointwise Wilson 95% interval at each trial count; it is not a simultaneous confidence band.

    Running estimate vs trials
    TrialsEmpiricalTheoryPointwise Wilson 95% interval

    Teaching tips

    FAQ

    How can I reproduce the same simulation?

    Keep the experiment, parameters, trial count, and integer seed the same. This page uses the same LCG to reproduce the same sequence, and the scenario link shares those settings. The generator is for teaching, not cryptographic use.

    Why does the empirical result differ from the theoretical probability?

    Finite samples vary by chance. Increase the number of trials and inspect the running chart and its pointwise Wilson 95% interval. The shaded region is not a simultaneous confidence band for the whole path.

    What is included when I export or share a simulation?

    A scenario link contains only the experiment settings and seed. CSV, LaTeX, and the text summary contain the result for the current revision. Run again after changing a setting; exports made while paused are marked as partial.