Conditional probability
P(A then B) = P(A) × P(B | A)
P(B | A) is the conditional probability of B after A has already happened. Without replacement, that second probability usually changes because one item has been removed.
Learn why each draw changes the next probability, then solve a tree one branch at a time.
Without replacement means that the item you draw stays out of the collection. The next draw is made from what remains, so its probability depends on the first draw.
That is why a tree diagram without replacement uses conditional probabilities on later branches instead of repeating the first set of fractions.
P(A then B) = P(A) × P(B | A)
P(B | A) is the conditional probability of B after A has already happened. Without replacement, that second probability usually changes because one item has been removed.
Start with the full collection. Draw one branch for every possible first result, and write each probability as number of that result ÷ total items.
Treat each first-draw branch separately. Remove the item drawn on that branch, so the total decreases by one and that result’s count may decrease too.
From each updated collection, draw every possible second result. Use the remaining total and remaining counts when writing these probabilities.
Multiply probabilities along a complete route. When the same event can happen in several different orders, add the probabilities of all matching mutually exclusive routes.
Guided walkthrough
Use the four steps above to build the tree and solve the question one action at a time.
A bag has 3 red and 2 blue marbles. Draw two marbles without replacement. What is the probability of one red and one blue?
Start: 3 red, 2 blue — 5 total.
Worked example
Problem
A box contains 4 black counters and 3 white counters. Two counters are drawn one after another without replacement.
Find
Find the probability of drawing exactly one black counter and one white counter.
Worked example
Problem
A class has 3 girls and 2 boys. Choose a president and then a vice-president without replacement.
Find
Find the probability that the president is a girl and the vice-president is a boy.
Worked example
Problem
A standard 52-card deck contains 4 aces and 48 non-aces. Draw two cards one after another without replacement.
Find
Find the probability of drawing at least one Ace.
Without replacement, one item is gone after the first draw. The second denominator must be one less than the first.
On each branch, reduce the count only for the category drawn on that branch before writing the next probability.
For one of each type, include both possible orders unless the question specifies an order, such as first then second.
For at least one target item, the complement is often simpler: find the chance of no target items, then subtract from 1.
| Topic | With replacement | Without replacement |
|---|---|---|
| Returned item | Returned to the collection | Removed from the collection |
| Second-draw denominator | Same as the first draw | One less than the first draw |
| Event relationship | Independent | Dependent |
| Calculation | Probabilities stay the same | Use probabilities conditional on the first draw |
A box has 5 black and 3 white counters. Draw two without replacement. Find P(two black).
5/8 × 4/7 = 5/14
From 4 seniors and 2 juniors, choose a captain then a vice-captain. Find P(exactly one junior).
4/6 × 2/5 + 2/6 × 4/5 = 8/15
A bag has 2 gold and 6 silver tokens. Draw two without replacement. Find P(at least one gold).
1 − (6/8 × 5/7) = 13/28
Quick answers about dependent draws and probability trees.
Divide the number of wanted objects by the current total, remove the drawn object, then multiply the matching path probabilities.
This page uses counters, class representatives, and a standard deck to show one path, paths in either order, and a complement.
With replacement restores the collection so later events are independent. Without replacement changes the collection so later events are dependent.
The first object is no longer available, so one fewer object remains for the second draw.
Use the calculator to model your next probability question.