What Mutually Exclusive Events Mean

When Two Events Cannot Happen Together

In Mixed Practice on Basic Probability Rules, you used the addition rule for “A or B” and accounted for outcomes that belong to both events. This tutorial focuses on the special case where there are no shared outcomes. For example, on one roll of a number cube, the result can be even or it can be 5—but it cannot be both.

Whether events can happen together depends on the chance process and what counts as one complete outcome. We are asking whether the same outcome from one trial can satisfy both event descriptions. Two events might occur on different trials without any conflict; “mutually exclusive” describes what can happen together on a single trial.

The notation makes the definition precise: \(A\cap B\) is the event that both \(A\) and \(B\) occur. If the intersection contains no outcomes, the events are disjoint. A practical way to check is to ask, “Can I name one complete outcome that satisfies both event descriptions?” If you can, the events overlap. If you cannot, they are disjoint in the stated model.

The word “complete” matters. In a multi-stage process, an outcome must include all the recorded stages. For a roll followed by a coin flip, for instance, an outcome is a pair such as (2, heads), not just the number or just the coin result. The events must be checked against the same complete outcome.

What Disjointness Does to an “Or” Probability

In Choosing Between Counting and Formula Approaches, you used the general addition rule: add the probabilities of two events and subtract the probability of their overlap. When events are disjoint, there is no overlap to subtract. This gives a useful special case.

Here, \(A\cup B\) means that \(A\) occurs, \(B\) occurs, or both occur. For disjoint events, “or” cannot mean both on the same trial, because there are no outcomes in both events. The special addition rule is not a different meaning of “or”; it is the general rule applied when the intersection is empty.

It is useful to keep two related statements separate. Disjointness is a statement about outcomes: \(A\cap B=\varnothing\). It follows that \(P(A\cap B)=0\). But in some probability models, an intersection can contain an outcome and still have probability zero. So a zero intersection probability does not always prove that two events are disjoint.

For finite sample spaces where every outcome has positive probability, the converse does hold: if \(P(A\cap B)=0\), then \(A\cap B\) must be empty. Later probability models can include individual outcomes with probability zero, so the safer definition is always based on whether the events share outcomes.

The events are not disjoint because they are unlikely to happen together; they are disjoint because no possible result can make both descriptions true. In this example, the result 5 belongs to \(F\), while the even results belong to \(E\), and the sets do not overlap.

Worked Example: Red or a King

This example shows why an event pair should not be labeled disjoint just because its descriptions sound different. “Red” and “king” describe different properties of a card, but some cards have both properties. In contrast, a card cannot be both a heart and a spade, so the events “the card is a heart” and “the card is a spade” are disjoint for one card selection.

Worked Example: Mutually Exclusive Categories in a Kiosk

That qualification is important in everyday settings. “A visitor uses the bus” and “a visitor carries a backpack” could both be true for the same visitor. But if a table assigns each trip exactly one arrival method, “the trip is recorded as bus” and “the trip is recorded as car” cannot both be true. Do not decide based only on the labels; check what one outcome records and whether both descriptions can apply to it.

Disjoint Events Are Not the Same as Independent Events

The terms disjoint and independent describe different ideas. Disjoint events cannot occur together on one trial. Independence concerns whether information about one event changes the probability of the other. The earlier tutorial The Gambler’s Fallacy and Independent Trials discussed independence in repeated trials; do not use “independent” as another word for “disjoint.”

For example, on one roll of a fair number cube, “the result is even” and “the result is 5” are disjoint. Learning that the result is even rules out a 5 on that roll. In ordinary cases where both events have positive probability, disjoint events are not independent. For this tutorial, the key distinction is simple: disjointness asks whether the events share outcomes; independence asks a different probability question.

Common Mistakes and AP Exam Tips

• Checking whether events sound different instead of checking outcomes. “Red” and “king” sound like different descriptions, but a red king satisfies both. List or describe the outcomes that meet each event and check for a shared one.
• Thinking “or” always means the events are disjoint. “Or” describes the union; it does not say whether there is overlap. First check the intersection, then decide whether the general addition rule or its disjoint-events special case applies.
• Adding probabilities without checking the overlap. Add \(P(A)\) and \(P(B)\) directly only when the events are disjoint. If they overlap, use the general addition rule and subtract \(P(A\cap B)\) once.
• Calling events disjoint merely because their intersection probability is zero. A probability-zero intersection need not be an empty set in every model. Define disjointness by \(A\cap B=\varnothing\); in a finite sample space with positive probability for every outcome, zero intersection probability does imply an empty intersection.
• Confusing different trials with the same trial. Two events can occur on separate repetitions even if they cannot occur together on one repetition. State what one trial records before deciding whether the events are disjoint.
• Leaving the justification implicit. A full-credit response names the events, identifies whether any outcome belongs to both, and connects that finding to the calculation. For disjoint events, say that the intersection is empty, so their probabilities can be added for the union.

A concise exam response for the number-cube example could say: “\(E=\{2,4,6\}\) and \(F=\{5\}\), so \(E\cap F=\varnothing\). The events are mutually exclusive, and \(P(E\cup F)=3/6+1/6=2/3\).” This states the evidence for disjointness and uses the appropriate rule.