Loading…
One experiment draws cards many times and estimates P(at least one target card in a draw). We then repeat the whole experiment many times to see how much the estimate itself varies.
Whole number from 1 to 52.
The same seed and settings always reproduce exactly the same results.
If an event has true probability p and an experiment uses m independent trials, the estimate p̂ = (successes ÷ m) has standard deviation:
SE = √( p (1 − p) / m )Because m sits under a square root, you need four times as much data to halve the typical error. This is why polls, clinical trials and simulations all report their sample size. The “Compare experiment sizes” button demonstrates it with 100, 1,000 and 10,000 draws per experiment.
The simulator also builds a 95% interval around each experiment's estimate. Counting how often those intervals contain the true probability checks that the interval method is honest: the share should be close to 95%, not 100% and not 50%. Read more in the sampling error lesson.