Skip to content

Advertisement

Negative binomial distribution calculator

Find the probability of failures before the r-th success or total trials to that success. See the CDF, mean and variance.

Probability of exactly this count: —

Fill in the fields above — the result is computed on this page.

P(K ≤ k), up to this many failures
—
Failures k
—
Trials t = k + r
—
Expected failures E(K)
—
Expected trials E(T)
—
Variance Var(K) = Var(T)
—
What these numbers mean
  • We compute it from the formula shown below — not from a prediction model and not from draw history.
  • r = 3, p = 0.5, k = 2: exactly 18.75%, up to two failures 50%.
  • In trial mode, enter at least r. The calculator subtracts r to find failures.

Trials must be independent with a constant p greater than zero. r: 1–1000, x: 0–10,000. Very small probabilities may round to zero. This model does not predict the next lottery draw.

How to use this page
  1. Load an example or enter your own data.
  2. Choose the calculation type and units where available.
  3. Read the result, additional measures and substituted formula.

What this page does not do: Trials must be independent with a constant p greater than zero. r: 1–1000, x: 0–10,000. Very small probabilities may round to zero. This model does not predict the next lottery draw.

Further reading

Formula

P(K=k) = C(k+r−1,k) pʳ (1−p)ᵏ

Repeat independent trials with a constant success probability p until r successes occur. K counts failures before the last success. Total trials T = K + r includes successes. The last trial must be a success.

For r = 3, p = 0.5 and k = 2, C(4,2) × 0.5³ × 0.5² = 0.1875, or 18.75%. The probability of at most two failures is 50%. Selecting five total trials gives the same point probability.

A binomial model fixes the number of trials and counts successes. A negative binomial model fixes successes and counts the trials or failures needed. For r = 1 it is the geometric distribution.

Frequently asked questions

How can I check the example?

r = 3, p = 0.5, k = 2: exactly 18.75%, up to two failures 50%.

What should I check before interpreting the result?

Trials must be independent with a constant p greater than zero. r: 1–1000, x: 0–10,000. Very small probabilities may round to zero. This model does not predict the next lottery draw.

See also

← All calculators

Advertisement
Offline app and draw notifications

My saved selections

Selections saved only in this browser. Numbers are shown exactly as saved; this is not a lottery ticket.

Open the My saved selections hub
Restore with preview

Restore in the hub: review the file and confirm first. Both old backups and the complete selections + journal backup are supported. Nothing is uploaded.