Chapter 3

Probability Fundamentals

📖 ~50 min read 📈 1 interactive chart ✍️ 3 practice questions 🎯 1 linked game

1. What Is Probability?

Probability is the mathematical language of uncertainty. Every business decision involves some degree of the unknown — will a shipment arrive on time? Will a customer return a product? Will demand exceed supply this quarter? Probability gives us a framework for quantifying these uncertainties and making rational decisions in the face of incomplete information.

At its core, a probability is a number between 0 and 1 (or equivalently, between 0% and 100%) that expresses how likely an event is to occur. A probability of 0 means the event is impossible; a probability of 1 means it is certain. Most events of business interest fall somewhere in between.

Three Approaches to Probability

There are three main ways to assign probabilities to events:

  • Classical probability: Based on equally likely outcomes. If you roll a fair die, each face has a probability of 1/6. This approach works well when the sample space is known and outcomes are symmetric.
  • Empirical probability: Based on observed data. If 162 out of 200 deliveries arrived on time, the empirical probability of on-time delivery is 162/200 = 0.81. This is the most common approach in business analytics.
  • Subjective probability: Based on expert judgment or personal belief. A supply chain manager might estimate a 70% chance that a new supplier will meet quality standards, even without historical data. This approach is used when data is scarce or unavailable.

Sample Space and Events

The sample space (denoted S) is the set of all possible outcomes of an experiment or process. An event is any subset of the sample space — it is the outcome or set of outcomes we are interested in. The complement of an event A (denoted A′) consists of all outcomes in the sample space that are not in A.

For example, if LakeFront Retail tracks whether each delivery is on time or late, the sample space is S = {On Time, Late}. The event A = {On Time} has complement A′ = {Late}.

🏪 LakeFront Retail Co.

LakeFront Retail Co. receives hundreds of shipments from suppliers each month. Management wants to understand delivery reliability to negotiate better contracts and set inventory buffers. They begin by analyzing last month's delivery data: out of 200 deliveries, 162 arrived on time and 38 were late.

This delivery data will serve as our working example throughout this chapter as we explore the fundamentals of probability.

Probability of an Event
📊 Excel: =COUNTIF(range,criteria)/COUNT(range)
where P(A) is the probability of event A, and the ratio counts favorable outcomes over total outcomes.
Complement Rule
📊 Excel: =1-COUNTIF(range,criteria)/COUNT(range)
The probability that event A does not occur equals 1 minus the probability that it does occur.
✎ Worked Example: On-Time Delivery Probability
1
Identify the data:
Total deliveries = 200   On-time deliveries = 162
2
Calculate P(On Time):
P(On Time) = 162 / 200 = 0.81
3
Calculate P(Late) using the complement rule:
P(Late) = 1 − P(On Time) = 1 − 0.81 = 0.19
4
Result: LakeFront can expect about 81% of deliveries to arrive on time and 19% to be late. This 19% late rate may justify increasing safety stock for critical items.
✓ Check Your Understanding
If P(on-time delivery) = 0.81, what is P(late delivery)?
0.81
0.19
0.09
1.81
💡 Key Takeaway

Probability always falls between 0 and 1, and the probabilities of all possible outcomes must sum to 1. The complement rule is one of the most useful shortcuts in probability: if you know P(A), you immediately know P(not A) = 1 − P(A).

2. Addition and Multiplication Rules

Most real business questions involve combinations of events. Will Supplier A or Supplier B be late? Will both shipments arrive on time? To answer these questions, we need the addition rule and the multiplication rule.

Mutually Exclusive Events

Two events are mutually exclusive (disjoint) if they cannot both occur at the same time. For example, a single delivery cannot be both on time and late — these events are mutually exclusive. When events are mutually exclusive, P(A and B) = 0, which simplifies the addition rule.

Independent Events

Two events are independent if the occurrence of one does not affect the probability of the other. For example, if LakeFront orders from two different suppliers, whether Supplier A is late does not affect whether Supplier B is late (assuming they operate independently). Independence simplifies the multiplication rule.

Addition Rule (General)
📊 Excel: =P_A + P_B - P_A_and_B
where P(A or B) is the probability that at least one of the events occurs. Subtract P(A and B) to avoid double-counting outcomes in both events.
Multiplication Rule (Independent Events)
📊 Excel: =P_A * P_B
For independent events, the probability that both occur equals the product of their individual probabilities.
🏪 LakeFront Retail Co.

LakeFront orders inventory from two independent suppliers. Supplier A is late 12% of the time (P = 0.12). Supplier B is late 18% of the time (P = 0.18). Since the suppliers operate independently, management wants to know: what is the probability that both suppliers are late on the same order cycle?

✎ Worked Example: Both Suppliers Late
1
Identify the given probabilities:
P(A late) = 0.12   P(B late) = 0.18
2
Confirm the events are independent (the suppliers operate separately).
3
Apply the multiplication rule for independent events:
P(A late and B late) = P(A late) × P(B late)
P(both late) = 0.12 × 0.18 = 0.0216
4
Result: There is approximately a 2.16% chance that both suppliers are late in the same cycle. While individually each supplier has a notable late rate, the probability of both being late simultaneously is quite small — good news for LakeFront's inventory planning.
✓ Check Your Understanding
Two independent events each have a probability of 0.1. What is P(both occur)?
0.2
0.1
0.01
0.19
💡 Key Takeaway

The addition rule handles “or” questions (at least one event occurs), while the multiplication rule handles “and” questions (both events occur). For independent events, the multiplication rule simplifies to just multiplying the individual probabilities. Always check whether events are mutually exclusive or independent before applying these rules.

3. Conditional Probability and Bayes’ Theorem

In many business situations, the probability of an event changes when we learn new information. If a product is defective, which supplier likely produced it? If a customer made a purchase last month, what is the probability they will buy again this month? These are questions of conditional probability — the probability of one event given that another has already occurred.

Conditional Probability

The conditional probability of A given B, written P(A | B), tells us the probability of event A occurring given that we already know event B has occurred. It is calculated by dividing the probability that both A and B occur by the probability of B.

Bayes’ Theorem

Bayes’ Theorem is one of the most powerful tools in probability. It allows us to reverse conditional probabilities — to update our beliefs about causes when we observe effects. If we know the probability of an effect given a cause, Bayes’ Theorem lets us compute the probability of a cause given the observed effect.

In a business context, this is invaluable. If we find a defective product on the shelf, Bayes’ Theorem helps us determine which supplier most likely produced it, based on each supplier’s defect rate and share of total supply.

Conditional Probability
📊 Excel: =COUNTIFS(range1,A,range2,B)/COUNTIF(range2,B)
where P(A | B) is the probability of A given that B has occurred.
Bayes’ Theorem
📊 Excel: =P_B_given_A * P_A / P_B
Bayes’ Theorem reverses conditional probabilities: given observed data B, update the probability of cause A.
🏪 LakeFront Retail Co.

LakeFront receives products from two suppliers. Supplier A provides 60% of products with a 3% defect rate. Supplier B provides 40% of products with an 8% defect rate. A quality inspector finds a defective item on the shelf. Which supplier is most likely responsible?

✎ Worked Example: Bayes’ Theorem — Identifying the Defect Source
1
Identify the given information:
P(Supplier A) = 0.60   P(Supplier B) = 0.40
P(Defect | Supplier A) = 0.03   P(Defect | Supplier B) = 0.08
2
Calculate the total probability of a defect using the law of total probability:
P(Defect) = P(Defect | A) × P(A) + P(Defect | B) × P(B)
P(Defect) = (0.03)(0.60) + (0.08)(0.40)
P(Defect) = 0.018 + 0.032 = 0.050
3
Apply Bayes’ Theorem to find P(Supplier A | Defect):
P(A | Defect) = P(Defect | A) × P(A) / P(Defect)
P(A | Defect) = (0.03)(0.60) / 0.050
P(A | Defect) = 0.018 / 0.050 = 0.36
4
Calculate P(Supplier B | Defect) as confirmation:
P(B | Defect) = (0.08)(0.40) / 0.050 = 0.032 / 0.050 = 0.64
5
Result: Even though Supplier A provides 60% of products, given a defect is found there is only a 36% chance it came from Supplier A and a 64% chance it came from Supplier B. Supplier B’s much higher defect rate (8% vs. 3%) outweighs its smaller share of supply.
Probability Tree — Supplier Defect Analysis
✓ Check Your Understanding
In the LakeFront example above, P(defect came from Supplier A) is approximately:
60%
36%
50%
75%
💡 Key Takeaway

Bayes’ Theorem lets you update probabilities when new information becomes available. The “prior” probability (Supplier A provides 60%) gets updated to a “posterior” probability (only 36% chance the defect came from Supplier A) once we observe the evidence. This is the foundation of data-driven decision-making.

🎮
Practice: Probability Puzzler Test your understanding of probability rules and Bayes’ Theorem

4. Counting Methods

Many probability problems require counting the number of ways an event can occur. If LakeFront needs to choose 3 suppliers out of 8 candidates, how many different groups of 3 are possible? Counting methods — factorials, permutations, and combinations — provide the tools to answer these questions systematically.

Factorials

The factorial of a positive integer n, written n!, is the product of all positive integers from 1 to n. Factorials grow extremely fast — 10! = 3,628,800 — and form the basis for both permutation and combination formulas. By convention, 0! = 1.

Permutations

A permutation counts the number of ways to arrange r items selected from n distinct items when order matters. For example, if LakeFront must rank its top 3 suppliers out of 8, the order of the ranking matters (1st, 2nd, 3rd), so we use permutations.

Combinations

A combination counts the number of ways to select r items from n distinct items when order does not matter. If LakeFront simply needs to choose 3 suppliers from 8 (without ranking), only the group membership matters, so we use combinations.

Factorial
📊 Excel: =FACT(n)
where n! is the product of all positive integers from 1 to n.
Permutations
📊 Excel: =PERMUT(n,r)
where n is the total number of items, r is the number selected, and order matters.
Combinations
📊 Excel: =COMBIN(n,r)
where n is the total number of items, r is the number selected, and order does not matter.
🏪 LakeFront Retail Co.

LakeFront is expanding its supplier network. The procurement team has identified 8 qualified candidates and needs to select 3 new suppliers. Since the selection is about which suppliers to partner with (not about ranking them), this is a combinations problem.

✎ Worked Example: Choosing Suppliers (Combinations)
1
Identify the values:
n = 8 (total candidates)   r = 3 (suppliers to choose)
2
Apply the combination formula:
C(8, 3) = 8! / (3! × (8 − 3)!)
3
Calculate the factorials:
8! = 40,320   3! = 6   5! = 120
4
Compute the result:
C(8, 3) = 40,320 / (6 × 120) = 40,320 / 720 = 56
5
Result: There are 56 different possible groups of 3 suppliers that LakeFront could select from the 8 candidates. In Excel, this is simply =COMBIN(8,3).
💡 Key Takeaway

Use combinations when order does not matter (selecting suppliers, choosing products for a promotion, forming a committee). Use permutations when order matters (scheduling deliveries, ranking candidates, assigning tasks to time slots). The key question to ask: does rearranging the same items produce a different outcome? If yes, use permutations. If no, use combinations.

📄
Download the Probability Practice Workbook
Coming Soon — Excel file with delivery data and probability exercises

5. Chapter Summary

In this chapter, we covered the foundational concepts of probability — the mathematical framework for reasoning about uncertainty. Here is what you should take away:

💡 Chapter 3 Summary

Basic Probability: Probability quantifies uncertainty on a 0-to-1 scale. The complement rule (P(A′) = 1 − P(A)) is one of the most frequently used shortcuts.

Addition and Multiplication Rules: The addition rule handles “or” scenarios; the multiplication rule handles “and” scenarios. For independent events, P(A and B) simplifies to P(A) × P(B).

Conditional Probability & Bayes’ Theorem: Conditional probability updates our beliefs when new information arrives. Bayes’ Theorem is the formal tool for reversing cause and effect — going from P(effect | cause) to P(cause | effect).

Counting Methods: Factorials, permutations, and combinations provide systematic ways to count outcomes. Use permutations when order matters; use combinations when it does not.

📋 Chapter 3 — Formula Reference
Concept Formula Excel Function
Probability
=COUNTIF()/COUNT()
Complement
=1-P(A)
Addition Rule
=P_A+P_B-P_A_and_B
Multiplication (Indep.)
=P_A*P_B
Conditional
=COUNTIFS()/COUNTIF()
Bayes’ Theorem
=P_B_given_A*P_A/P_B
Factorial
=FACT(n)
Permutations
=PERMUT(n,r)
Combinations
=COMBIN(n,r)
Up Next
Chapter 4: The Normal Distribution