Find the chance of exactly, at most or at least k events when they happen at random at an average rate λ, like calls per hour or defects per batch.
The Poisson formula
P(X = k) = e^(−λ) × λ^k ÷ k!
Mean = λ · Variance = λ · SD = √λ
λ (lambda) is the average number of events in the period you care about, and k is a count: 0, 1, 2 and so on. The cumulative answers add up the terms. Each term is worked out with logarithms, so large rates, like λ = 1,000, don't overflow.
Worked example
A help desk gets an average of 3 calls an hour, at random. The chance of exactly 2 calls in an hour is e^(−3) × 3² ÷ 2! = 0.049787 × 9 ÷ 2 = 0.224042, about 22.4%.
The chance of 2 or fewer is e^(−3) × (1 + 3 + 4.5) = 0.423190. The chance of 5 or more is 1 − P(X ≤ 4) = 0.184737, about 18.5%. The chance of no calls at all is e^(−3) = 0.0498.
| Calls in an hour (k) | P(X = k) | P(X ≤ k) |
|---|
| 0 | 0.0498 | 0.0498 |
| 1 | 0.1494 | 0.1991 |
| 2 | 0.2240 | 0.4232 |
| 3 | 0.2240 | 0.6472 |
| 4 | 0.1680 | 0.8153 |
| 5 | 0.1008 | 0.9161 |
| 6 | 0.0504 | 0.9665 |
| 7 | 0.0216 | 0.9881 |
| 8 | 0.0081 | 0.9962 |
To change the period, scale λ. 3 calls an hour is 1.5 calls per half hour, or 24 calls in an 8-hour shift.
When the Poisson model fits
- Events happen one at a time and independently of each other.
- The average rate stays the same across the period. Rush hours and quiet times break this, so model them separately.
- You're counting events in a fixed stretch of time, distance or area, with no fixed number of tries.
- If the data spread out more than the mean suggests (variance much bigger than the mean), the real chance of extreme counts is higher than Poisson says.
Sources
How we checked this
Before every update, this calculator is run in a real browser with these known answers, and it must match all of them:
- λ = 3: P(X = 2) = 0.224042 ✓
- λ = 3: P(X ≤ 2) = 0.423190 ✓
- λ = 3: P(X ≥ 5) = 0.184737 ✓
- λ = 3: P(X = 0) = e^−3 = 0.049787 ✓
- Large rate: λ = 1,000, P(X = 1,000) = 0.012615 ✓
- Edge: a rate of 0 is rejected ✓
Last checked October 26, 2026. Spotted something wrong? The math is done in your browser with the formula shown on this page, so you can check any result by hand.
Questions people ask
What is the Poisson distribution used for?
Counting random events in a fixed period or space: calls per hour, typos per page, accidents per month, or customers per minute.
How do you calculate a Poisson probability?
P(X = k) = e^(−λ)λ^k ÷ k!. With λ = 3 and k = 2: e^(−3) × 9 ÷ 2 = 0.2240.
What does λ mean?
The average number of events in the period. It's also the variance, so the standard deviation is √λ.
How do I find the probability of at least one event?
1 − P(X = 0) = 1 − e^(−λ). With λ = 3 that's 1 − 0.0498 = 0.9502.
What's the difference between Poisson and binomial?
Binomial counts successes in a fixed number of tries with the same chance each time. Poisson counts events in a period with no fixed number of tries. Binomial with many tries and a small chance is close to Poisson.