Poisson Distribution Calculator

Exact Poisson probabilities P(X = k), cumulative P(X ≤ k), P(X ≥ k), plus mean, variance and standard deviation — with a full distribution chart and step-by-step working.

The Poisson distribution explained, step by step

The Poisson distribution models the number of independent events that occur in a fixed interval when the average rate λ is constant. Each card below covers one piece of the formula this calculator uses.

1. λ is a rate — a count per interval, not a probability

λ (lambda) is the average number of events you expect over a fixed window of time, area or volume. It can be any positive real number — not bounded by 1 — and it scales with the window: doubling the interval doubles λ.

X ~ Poisson(λ)
λaverage events per interval·Xactual count observed
Example
Given
12 customers/hr, 30-min window
Substitute
λ = 12 · 0.5
Answer
λ = 6

2. The PMF — probability of exactly k events

λ^k weighs scenarios producing k events, e^(−λ) normalises the distribution so probabilities sum to 1, and dividing by k! removes the double-count from arrival orders. The calculator uses log-gamma so k! stays stable even for large k.

P(X = k)
=k · e−λ) / k!
eEuler's number ≈ 2.71828·k!k factorial
Example
Given
λ = 5
k = 3
Substitute
125 · 0.006738 / 6
Answer
P ≈ 0.1404

3. Mean = variance = λ (signature property)

For any Poisson variable the mean and variance are both λ, so σ = √λ. If real data has variance much larger than its mean (over-dispersion) or much smaller, Poisson is the wrong model — that is the standard diagnostic.

μ = λ   σ²
= λ   σ
= √λ
Example
Given
λ = 9
Substitute
μ = 9, σ²
= 9
Answer
σ = 3

4. When to switch models — binomial, exponential, normal

Fixed number of trials → binomial (Poisson is the limit as n → ∞, p → 0 with np = λ). Waiting time between events → exponential with rate λ — same process, different question. Once λ ≥ 20, Poisson(λ) is well approximated by Normal(λ, λ) with a continuity correction.

large n
small p → Poisson(np) · large λ → Normal(λ, λ)
counts / interval → Poisson
gaps between events → Exponential(λ)
λ ≥ 20 → Normal(λ, λ)
Example
Given
n = 1000
p = 0.003
Substitute
λ = np
= 3
Answer
≈ Poisson(3)

Features of this calculator

  • Exact Poisson probabilities: P(X = k), P(X ≤ k), P(X < k), P(X ≥ k), P(X > k) — all in one calculation.
  • Uses log-gamma internally, so probabilities stay accurate even when k! is astronomically large.
  • Probability bar chart automatically spans a meaningful range of k values (roughly λ ± 4·√λ), with the bar for your chosen k highlighted.
  • Mean, variance and standard deviation reported alongside the probabilities — with the useful reminder that mean = variance = λ.
  • Show/hide step-by-step working with the formula, λ^k, e^(−λ) and k! all laid out.

Frequently asked questions

+What if I don't know λ?

Estimate it from data as the sample mean (total events ÷ number of intervals). If a call centre received 1,200 calls over 100 hours, an estimate of λ per hour is 12.

+How large can k be in practice?

Mathematically k has no upper limit. In practice P(X = k) becomes negligibly small once k is several standard deviations above λ. This calculator draws the chart out to about λ + 4·√λ, which covers essentially all the probability mass for reasonable λ.

+Can λ be 0?

Yes, but the distribution collapses: P(X = 0) = 1 and P(X = k) = 0 for any k ≥ 1. Not very useful in modeling terms.

+When can I use a normal approximation?

Once λ is large (a common threshold is λ ≥ 20), Poisson(λ) is well approximated by Normal(μ = λ, σ² = λ). Use a continuity correction if you approximate discrete probabilities from a continuous distribution.

+How is Poisson related to the exponential distribution?

If events happen according to a Poisson process with rate λ, the time between consecutive events follows an exponential distribution with rate λ. They're two views of the same underlying process.

Related calculators