Counts have their own distributions:
- Binomial: the number of successes in n independent tries, each with probability p, such as trials that convert to members.
stats.binom.pmf(k, n, p) is the probability of exactly k.
- Poisson: the number of events in an interval when they arrive independently at a steady rate, such as bikes rented in an hour.
stats.poisson.pmf(k, lam) with lam the average count.
For "at least" and "more than", use the survival function sf, which gives the probability above a value: sf(k) is P(X > k). So "at least 5" is sf(4). This off-by-one is the most common mistake with counts.
A binomial count has mean n × p and SD sqrt(n × p × (1 - p)).