How can a probability tree help me keep multi-step cases straight?
Quick answer
A probability tree lays out each possible outcome of a sequence as branches, with the probability of each outcome written on its branch. To find the probability of one complete path, multiply its branch probabilities; to find an event that can happen along several distinct paths, add those path probabilities. Later branches can have different probabilities depending on what happened earlier, so label them accordingly. [S1][S2]
The question, the details, the next step.
I keep losing track of what can happen when an event has two steps. How does a probability tree help?
A tree gives each step its own layer. Start with the first possible outcomes, then draw a new branch from each outcome for the next step. Label branches with their probabilities, and read from the start of the tree to a final outcome to get one complete case. A tree represents possible outcomes and their associated probabilities or frequencies; it organizes the cases rather than suggesting they are equally likely. [S2]
The key is to label each later branch for the situation you are actually in. Its probability may depend on the earlier outcome.
Could you show me a small tree where the second step changes depending on the first?
Assume, just for this example, a bag contains two red counters and one blue counter. You draw two counters in order without putting the first one back. The branch labels below are probabilities under that assumption:
Start
├─ Red (2/3)
│ ├─ Red (1/2) → Red, then Red
│ └─ Blue (1/2) → Red, then Blue
└─ Blue (1/3)
└─ Red (1) → Blue, then RedThere is no Blue-then-Blue path: after drawing the only blue counter first, none is left. The second-step branches therefore depend on the first draw. This is a hypothetical model to illustrate the method, not a claim about a real draw.
How do I use that tree if I want the chance of getting one red and one blue, in either order?
First identify the complete paths that match the event. Here they are Red-then-Blue and Blue-then-Red. Multiply along each path: Red-then-Blue is 2/3 × 1/2 = 1/3; Blue-then-Red is 1/3 × 1 = 1/3. Since these are distinct paths that both match the event, add them: 1/3 + 1/3 = 2/3. For a sequential tree, multiplying along an ordered path and adding the probabilities of distinct matching paths is the reading method. [S1]
Keep the order in each path visible until you have identified which paths count. If the question asks for a particular order, only use that path.
What should I check before trusting my answer, especially if the branches are not equally likely?
Write down the setup first: what the possible outcomes are, whether the first result changes what is possible next, and whether anything is replaced or repeated. Then label each branch for its specific point in the tree, not with a probability copied from a different branch. In this example, the first draw changes the contents of the bag, so the second-draw probabilities differ by first-draw outcome.
A practical check is that the outgoing branch probabilities at a decision point should account for the possible next outcomes in your stated model. Here, after Red they are 1/2 and 1/2; after Blue the only possible next draw is Red with probability 1. The tree method organizes and combines the cases, but it does not establish whether your starting assumptions match a real situation. Check those assumptions separately.
A practical checklist
State the sequence and the assumptions, including whether an outcome is replaced or repeated.
Draw a branch for each possible outcome at every step.
Label each branch with the probability for that specific point in the tree.
Trace complete paths and select only the ones matching the event.
Multiply along each selected path, then add probabilities for distinct matching paths.
A few more useful answers
Do all branches in a probability tree have to be equally likely?
No. Each branch is labeled with the probability for that outcome at that point in the sequence. A later branch can have a different probability depending on the earlier outcome. [S1]
When should I add probabilities instead of multiplying them?
Multiply the branch probabilities along one complete path. If an event is represented by multiple distinct paths, add the probabilities of those paths. [S1]
What if the event I care about is only one possible outcome?
Trace the single path that describes it and multiply its branch probabilities. Make sure the path’s order matches the question.
Sources
References used for this guide.
- [S1] Tree and Venn Diagrams — Introductory Statistics openstax.org · Checked
- [S2] Chapter 3 Key Terms — Introductory Statistics openstax.org · Checked