What You Will Be Able to Do
By the end of this lesson, you should be able to:
- Compute
by restricting to B and counting - Explain why conditioning makes B the new sample space
- Connect the count
to the formula
Given It's Greater Than 3, Is It Even?
Roll a fair die. The outcomes are
- Let
= "greater than 3" - Let
= "even"
Given that the roll is greater than 3, what's the chance it's even? Try counting — don't reach for a formula yet.
Throw Away Every Outcome Not in B
Only
Conditioning Makes B the New Sample Space
When you condition on B, you replace the sample space with B.
- The roll is now one of
— three outcomes - You count A only among those three
- The denominator becomes
, not the original total
Compare: 2/3 Is Not 3/6
| Question | Outcomes counted | Probability |
|---|---|---|
| all 6 | ||
| just |
Same event "even" — but conditioning changed the denominator from 6 to 3.
What Is the New Denominator Now?
You condition on
After restricting to B, how many outcomes are you counting among?
Think before advancing — what number goes on the bottom?
Write the Move as a Counting Fraction
You just did this: count A among B's survivors, divide by how many survived.
The top is the overlap — outcomes in both A and B. The bottom is all of B.
The Counting Rule for Uniform Models
In a model where every outcome is equally likely:
— how many outcomes are in both A and B — how many outcomes are in B
Why It Equals the Formula You Know
Divide top and bottom by the total count
The total
Cards: P(king | face card) = 1/3
- Among the 12 face cards (B), 4 are kings (A)
Verify the Count With the Formula
The formula must give the same answer:
The 52s cancel — exactly the cancellation from two slides ago.
The Numerator Is the Overlap Only
For
- But 2 is not in B — it's not greater than 3
- Count only the evens that survived:
- Numerator is
, not
Your Turn: Restrict and Count
A fair die is rolled. You're told the result is odd.
What is
Restrict to the odds, count how many are less than 4, then form the fraction.
Two Common Errors to Watch For
Wrong denominator: dividing by the whole sample space, not by
Conditioning shrank the world to B — divide by B's count.
Wrong numerator: counting all of
An A-outcome outside B (like the 2) doesn't survive — don't count it.
Restrict to B, Then Count A
✓ Conditioning makes B the new sample space
✓
✓ It equals
Next: the same move across lists, Venn diagrams, and tables.
Click to begin the narrated lesson
Find conditional probability from a model