Statistics calculation guides

Calculate probability without replacement

Update the sample space after each draw, distinguish ordered and unordered outcomes, and verify probabilities stay within bounds.

Intent
Use this guide when successive events are dependent because the first selected item is not returned before the next draw.
Reviewed
Reading time
7 minutes

What this guide helps you decide

Without replacement, the first draw changes both the number of favorable items and the total available for the second draw. Treating the draws as independent repeats the original probability and generally produces the wrong result.

The event definition also matters. ‘Two red items’ is not the same as ‘one red then one blue,’ and an unordered pair may combine multiple ordered paths.

Update each conditional probability

For a specified sequence, multiply the probability of each step conditional on the earlier steps. For an unordered event, add the mutually exclusive ordered paths or use a matching combination model.

  1. Define whether order matters and whether selected items are replaced.
  2. Write the favorable and total counts before the first draw.
  3. Update both counts after each no-replacement outcome.
  4. Multiply along a path and add only mutually exclusive paths.
Worked scenario

Worked scenario: two red items

A bag contains five red and three blue items. Two items are drawn without replacement; find the probability both are red.

  1. First red: 5/8.
  2. After one red is removed, second red: 4/7.
  3. Both red: (5/8) × (4/7) = 20/56 = 5/14 ≈ 0.3571.

Outcome: The probability is 5/14, about 35.71%. Reusing 5/8 for the second draw would incorrectly assume replacement.

Probability-model checklist

  • Define the experiment and event precisely.
  • State replacement and ordering rules.
  • Update conditional counts after each draw.
  • Avoid double-counting overlapping outcomes.
  • Check that probabilities and their sums remain between zero and one.

Limits and responsible use

  • The result is only as valid as the assumed sample space and equal-likelihood model.
  • Observed processes can have bias or dependence not captured by a simple urn-style calculation.

Authoritative references

These links support the definitions, conventions, or safety boundaries used in this guide. CalculatorToolset wrote the explanation and example independently.

Frequently asked questions

When can I multiply probabilities directly?

Multiply conditional probabilities along the same path. For independent events the conditional value equals the original probability, but no-replacement draws are dependent.

When should I add probabilities?

Add probabilities for mutually exclusive paths that together make up the event, taking care not to count any outcome twice.