Why dice sums build a bell curve: convolution
Interactive 3D dice simulator. Roll up to five dice and watch the empirical sum histogram converge to the exact convolution probabilities: uniform to triangular to bell-shaped, the central limit theorem made visible.
Probability theory began with dice. In 1654 a gambler's questions about dice and unfinished games prompted a correspondence between Blaise Pascal and Pierre de Fermat that produced the first systematic calculus of chance. The questions they answered (how likely is each total, how should the stakes of an interrupted game be split) are the same expected-value calculations that price insurance and value portfolios today.
One die is as flat as randomness gets: every face has probability 1/6. Yet the sum of two dice is triangular, peaking at 7, and by five dice the distribution of the total is clearly bell-shaped. Nothing about an individual die changed; the shape comes from convolution, the operation that combines the distributions of independent quantities being added.
That is the central limit theorem in miniature, and it is why the normal distribution appears wherever many small independent effects add up: measurement errors, portfolio returns, aggregate claims. This module rolls one to five 3D dice (two by default), builds a histogram of the totals you roll, and overlays the exact convolution probabilities, so you can watch the observed frequencies close in on them as rolls accumulate.
Convolution of independent sums
P(S_n = s) = Σ_{k=1..6} P(S_{n−1} = s − k) · P(X = k)
S_n is the total of n dice and X is a single die, so P(X = k) = 1/6 for each face k from 1 to 6. Each added die convolves the previous total's distribution with that uniform one; repeated convolution pushes the shape toward the normal curve.
Frequently asked questions
Why is 7 the most likely total with two dice?
Because more ordered pairs sum to 7 than to any other total: six of the 36 equally likely outcomes (1+6 through 6+1), giving probability 6/36 ≈ 16.7%. Totals 2 and 12 have a single combination each, probability 1/36.
What is convolution in probability?
The operation that gives the distribution of a sum of independent random variables: the probability of each total is the sum over all ways of splitting it between the two addends. Adding a die convolves the running total's distribution with the single-die distribution.
How do dice demonstrate the central limit theorem?
The sum of independent, identically distributed variables, once standardised, approaches the normal distribution as their number grows. Dice show the mechanics at small n: uniform at one die, triangular at two, visibly bell-shaped by four or five. It is the same reason aggregate quantities in finance and insurance are often approximately normal.
Interactive explorer
The live model with sliders, charts and real-time statistics loads on this page. If it does not appear, enable JavaScript or open https://www.beyondaverages.org/dice in a modern browser.
Related modules
- Cards & hypergeometric: Interactive card-dealing simulator. Deal hands from a 3D deck and compare empirical frequencies of aces, pairs and colors against exact hypergeometric probabilities, and see why a deck's memory matters.
- Frequency: Interactive Poisson distribution explorer for insurance claim counts and hospital admissions. Move the rate λ and watch P(X=k), the mode, and the probability of zero or extreme counts update live.
- Severity: Interactive loss severity explorer comparing Log-Normal, Pareto (heavy-tail) and Gamma distributions. Adjust parameters and see the mean, median, 95th and 99th percentile claim sizes, and the share of total cost from the largest 5% of claims, update in real time.
- Expected loss: Interactive aggregate loss model for a personal motor book: Poisson claim frequency with Log-Normal severity across a portfolio of policies. Compute expected loss and run a 20,000-year Monte Carlo to see the aggregate loss distribution and its 95th, 99th and 99.5th percentiles.