Joint Probability Calculator
Calculate the probability that two events happen together using independent probabilities, conditional probability, a union probability, or a two-by-two table.
Find the overlap between two events
A joint probability calculator computes the probability that two or more events occur together. Calculate joint probability by multiplying probabilities when events are independent: P(A and B) = P(A) × P(B).
For dependent events, use P(A and B) = P(A) × P(B | A). Joint probability values range from 0 to 1, or from 0% to 100% when shown as a percentage.
Choose the method that matches what you know. Use independent events when one event does not change the other, conditional probability when one event is given, union probability when P(A or B) is known, or a count table when you have raw observations.
Results stay hidden until you click Calculate Joint Probability, so changed inputs never leave an old answer on screen.
Joint probability
Decimal: 0.0000
P(A)
--
P(B)
--
P(A or B)
--
Relationship
--
Calculation breakdown
Formula, inputs, and derived values.
Interpretation
How to read the result.
How to Use This Calculator
Start with the information you actually have. Different probability problems provide different pieces of the overlap.
1. Choose the method
Pick independent events, conditional probability, union probability, or a count table based on the data in your problem.
2. Enter probabilities or counts
Use percentages for probability inputs. For raw data, enter the four counts in the two-by-two table.
3. Calculate the overlap
Click Calculate Joint Probability to get P(A and B), its decimal form, and any derived values the inputs allow.
4. Check the interpretation
Use the relationship note to see whether the events are assumed independent, positively associated, negatively associated, or based on counts.
Joint Probability Formulas
Joint probability is the intersection of two events.
Independent events
P(A and B) = P(A) x P(B)
Conditional probability
P(A and B) = P(A) x P(B | A), or P(B) x P(A | B)
Union rule
P(A and B) = P(A) + P(B) - P(A or B)
Count table
P(A and B) = count where both happen / total count
Source
Penn State STAT 414 explains the multiplication rule used for joint probability in its Multiplication Rule lesson.
Worked Examples
These examples show which method to use and what the result means.
| Situation | Known values | Calculation | Joint probability |
|---|---|---|---|
| Independent exams | P(A)=80%, P(B)=60% | 0.80 x 0.60 | 48% |
| Customer behavior | P(A)=35%, P(B | A)=42% | 0.35 x 0.42 | 14.7% |
| Survey union | P(A)=45%, P(B)=35%, P(A or B)=65% | 0.45 + 0.35 - 0.65 | 15% |
Interesting Fact
Small overlaps can add up faster than intuition expects
A classic joint probability surprise is the birthday problem. Wolfram MathWorld reports that, assuming 365 equally likely birthdays and ignoring leap days, 23 people are enough for a shared-birthday probability of about 0.507297, or 50.7%. The reason is that 23 people create 253 pairwise comparisons, so many tiny overlap chances combine into one noticeably large event. See Wolfram MathWorld's Birthday Problem entry for the formula and statistic.
Translate Probability Wording Into Notation
Many joint probability errors start before the arithmetic. Use this guide to map common wording to the correct probability symbol and calculator method.
| Wording | Notation | Meaning | Calculator method |
|---|---|---|---|
| A and B | P(A and B) | Both events happen together. | Any method, depending on known values. |
| A or B | P(A or B) | At least one event happens, including cases where both happen. | Union method if P(A or B) is known. |
| B given A | P(B | A) | The probability of B inside the group where A already happened. | Dependent events: P(A) and P(B | A). |
| A given B | P(A | B) | The probability of A inside the group where B already happened. | Dependent events: P(B) and P(A | B). |
| Neither A nor B | 1 - P(A or B) | No event happens. | Count table or derived from a known union. |
The vertical bar in P(B | A) is the key detail: it means "given," not division. Read it as "B among the cases where A is true."
How to Check the Relationship Between Events
After you calculate the overlap, compare it with the overlap that independence would predict. This helps you describe whether the two events appear unrelated or associated.
Independence check
Compare P(A and B) with P(A) x P(B)
If the two values are the same, the events behave like independent events in the data. If the observed joint probability is higher, the events occur together more often than independence predicts.
Association direction
Higher or lower overlap changes the interpretation
A higher-than-expected overlap suggests positive association. A lower-than-expected overlap suggests negative association, and a zero overlap means the events are mutually exclusive in that sample or model.
| Comparison | Relationship | Plain-language interpretation |
|---|---|---|
| P(A and B) = P(A) x P(B) | Independent-like | Knowing A happened does not change the probability of B. |
| P(A and B) > P(A) x P(B) | Positive association | A and B happen together more often than independence would predict. |
| P(A and B) < P(A) x P(B) | Negative association | A and B happen together less often than independence would predict. |
| P(A and B) = 0 | Mutually exclusive | Both events cannot occur together in the defined experiment. |
Source: Penn State STAT 414 defines independent events and shows why independence implies P(A and B) = P(A) x P(B) in Two Definitions.
Build a Two-Way Table From Real Data
Raw observations are often easier to trust than copied percentages. A two-way table turns survey, product, experiment, or medical-screening records into probabilities the calculator can verify.
Define A and B clearly
Use binary definitions, such as subscribed or not subscribed, converted or not converted, tested positive or not positive.
Count each case once
Every record should fit exactly one of the four cells: both, A only, B only, or neither.
Check the total
The four cells should add to the full sample size after exclusions, missing data rules, and filters are applied.
Use counts directly
Enter the four counts into the table method instead of rounding percentages before calculation.
Keep the denominator consistent. A common mistake is mixing a percentage from one group with a count from another group, which can make P(A), P(B), and P(A and B) describe different populations.
Source: the NIST/SEMATECH e-Handbook describes the contingency table approach for cross-classified counts in How can we compare the results of classifying according to several categories?.
Which Method Should You Use?
Use the method that matches the way your problem is stated.
Use independent events when no event changes the other
This is common in idealized textbook problems, repeated random draws with replacement, or separate systems that do not influence each other.
Use conditional probability when wording says "given"
P(B | A) means the probability of B after A has occurred or after the data has been restricted to cases where A is true.
Use the union method when P(A or B) is known
The union rule prevents double-counting the overlap. If A or B is known, subtract it from P(A) + P(B).
Use counts when you have observations
A two-by-two table is often the clearest option for survey responses, medical tests, conversions, retention, defects, or categorical data.
Common Joint Probability Mistakes
Most wrong answers come from using the right-looking formula with the wrong assumption.
Assuming independence
Only multiply P(A) and P(B) directly when the problem says the events are independent or the assumption is justified.
Reversing conditional direction
P(B | A) and P(A | B) are usually different. Read the condition after the vertical bar as the group you are already inside.
Using impossible inputs
P(A or B) cannot be smaller than either event alone, and no probability can be below 0 or above 100%.
Probability questions
Frequently Asked Questions
Quick answers about joint probability, conditional probability, independence, unions, and count tables in probability theory and statistics.
What is joint probability?
Joint probability is the probability that two events happen together. It is written as P(A and B), and it represents the intersection between event A and event B inside the sample space. In plain terms, it is the likelihood that one combined outcome satisfies both event definitions at once.
How do I calculate joint probability for independent events?
Use the multiplication rule formula: P(A and B) = P(A) x P(B). For example, if P(A) is 80% and P(B) is 60%, the joint probability is 48%.
How do I calculate joint probability for dependent events?
Use conditional probability. If you know P(A) and P(B | A), multiply them: P(A and B) = P(A) x P(B | A). If you know P(B) and P(A | B), multiply those instead. Bayes theorem uses the same kind of conditional relationship when you need to reverse or update a probability after new information is known.
Can joint probability be greater than P(A) or P(B)?
No. The overlap cannot be larger than either event by itself, so P(A and B) must be less than or equal to both P(A) and P(B). Those single-event values are marginal probability values, while the joint value is limited to the part they share.
What is the difference between joint probability and union probability?
Joint probability means A and B happen together. Union probability means A or B happens, including cases where both happen. The union is usually larger than the joint probability.
Are mutually exclusive events independent?
Usually no. Mutually exclusive events cannot happen together, so their joint probability is 0. Independent events can still happen together; one simply does not change the probability of the other.
How does the count table method work?
Add all four cells to get the total sample size, then divide the count where A and B both happen by the total. When your data uses two categorical variables, the table shows the observed distribution across both variables and lets the calculator derive P(A), P(B), and P(A or B).
Why does the calculator reject some union inputs?
The union must be at least as large as the bigger individual probability and no larger than P(A) + P(B). If it falls outside that range, no valid overlap can produce those inputs.
Can this statistics calculator measure correlation or risk?
It can estimate overlap, association direction, or a simple risk likelihood when your inputs describe the same population. Treat the result as a decision aid for checking assumptions, not as proof of correlation, causation, or a complete statistical model.
Disclaimer
This calculator is for educational, statistical, and planning use. It assumes the probabilities or counts you enter are accurate and that the selected method matches the actual relationship between events.
The result is not a substitute for a full statistical model, study design, risk assessment, medical decision, financial decision, legal conclusion, or professional analysis. For high-stakes work, verify assumptions, sample quality, dependence structure, uncertainty, and confidence intervals with a qualified analyst.
Last updated: June 18, 2026