The controls, from top to bottom:
- Trials (n): 1 to 50 in steps of 1, starting at 10.
- Success probability (p): 0.00 to 1.00 in steps of 0.01, starting at 0.50.
- Exact, At most and At least buttons choose whether the live probability is P(X=k), P(X≤k) or P(X≥k).
- Outcome (k): any whole number from 0 to n. Its maximum follows the Trials slider, and if you lower n below k, k drops to n.
- Normal approximation button, Off or On.
The bar chart shows
P(X=k) for every
k from 0 to
n. Bars counted in your probability turn gold and a dashed line marks
k. Two boxes give
Mean μ = np and
SD σ = √(np(1−p)) to four decimal places, and the
Live probability box shows the answer, such as P(X ≥ 8) = 0.2277.
Switching the normal approximation on draws a teal normal curve with the same mean and SD over the bars, and a note under the probability. It reads
Normal approximation OK when
np≥10 and
n(1−p)≥10, and
Approximation unreliable otherwise. The curve is only a picture: the live probability is always the exact binomial value.
Watch the shape as you move
p. At
p=0.50 the bars are symmetric; below 0.50 they are skewed right, above 0.50 skewed left. Raising
n makes any shape more symmetric. With the normal curve on, at
n=40,
p=0.50 it traces the bars closely; at
n=10,
p=0.10 it fits badly.