A probability model says how a chance process should behave: a fair coin lands heads half the time. For a fixed number of independent tries, the probability of exactly k successes is the number of orders with k successes times p for each success and 1 - p for each failure. Listing every probability builds the distribution, whose expected value is n × p and whose spread is √(n p (1 - p)). A result within two standard deviations of the expected count is ordinary; a result farther out is surprising and makes the crew question the model.