Expected Value: The long-run average outcome of a random variable.
Net Winnings: Gross winnings minus cost to play.
Worth Playing: If E[Net Winnings] > 0, it's favorable to player; if < 0, it's unfavorable.
- Identify all possible outcomes and their probabilities
- Calculate net winnings for each outcome (gross - cost)
- Calculate expected value: E[Net Winnings] = Σ [net_winnings × P(outcome)]
- Make decision based on sign of expected value
Roll 6: P = 1/6, Net = $20 - $5 = $15
Roll 4 or 5: P = 2/6 = 1/3, Net = $10 - $5 = $5
Roll 1, 2, or 3: P = 3/6 = 1/2, Net = $0 - $5 = -$5
E[Net Winnings] = $15 × 1/6 + $5 × 1/3 + (-$5) × 1/2
= $2.50 + $1.67 - $2.50 = $1.67
Since E[Net Winnings] = $1.67 > 0, the game is favorable to the player.
On average, a player gains $1.67 per game played.
1/6 + 1/3 + 1/2 = 1/6 + 2/6 + 3/6 = 6/6 = 1 ✓
The expected net winnings are $1.67 per game.
This game is favorable to the player and worth playing.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Net Winnings: Gross winnings minus cost
• Game Evaluation: Positive EV favors player, negative favors house
Expected Return: The weighted average of all possible returns, weighted by their probabilities.
Investment Decision: Choose option with higher expected return (assuming risk neutrality).
E[A] = $10,000 × 0.7 + (-$5,000) × 0.3
= $7,000 - $1,500 = $5,500
E[B] = $15,000 × 0.5 + (-$3,000) × 0.5
= $7,500 - $1,500 = $6,000
E[A] = $5,500, E[B] = $6,000
Since $6,000 > $5,500, Investment B has the higher expected return
Investment B: Higher potential gain but also higher potential loss
Investment A: More conservative with lower variance
Investment B has the higher expected return of $6,000 compared to Investment A's $5,500.
Based on expected value alone, Investment B is the better choice.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Investment Comparison: Choose option with higher expected value
• Expected Value Dominance: Higher EV indicates better long-term outcome
Single Attempt: Binomial-like scenario with two outcomes.
Keep Trying Until Success: Geometric distribution scenario.
Net Reward: Points gained minus points spent.
E[Single] = 400 × 0.2 + (-100) × 0.8 = 80 - 80 = 0
This follows geometric distribution. Expected number of attempts until first success = 1/p = 1/0.2 = 5
Expected total cost = 5 × 100 = 500 points
Expected total reward = 500 points (received once)
Expected net = 500 - 500 = 0
E[Keep Trying] = Reward - (Expected Attempts × Cost per Attempt)
= 500 - (5 × 100) = 0
Both strategies have expected value of 0 points
The game is fair - neither advantageous nor disadvantageous
The expected value of a single attempt is 0 points.
The expected value of keeping trying until success is also 0 points.
This is a fair game with no advantage to either strategy.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Geometric Distribution: Expected attempts until success = 1/p
• Fair Game: Expected value of 0 indicates no advantage
Expected Value: The long-run average outcome of a random variable, representing the center of the probability distribution.
Net Winnings: Total earnings minus total costs in a game or investment.
Fair Game: A game where expected net winnings equal zero.
Risk-Neutral: Decision maker who focuses only on expected value, ignoring variance.
• Expected Value: E[X] = Σ [x × P(X = x)]
• Net Winnings: Gross - Costs
• Geometric Mean Trials: E[T] = 1/p
• Linear Transformation: E[aX + b] = aE[X] + b
Insurance Decision: Compare expected cost of risk vs. cost of insurance.
Expected Loss: Probability of event × Potential loss.
E[Loss without insurance] = $10,000 × 0.03 + $0 × 0.97 = $300
E[Cost with insurance] = $400 (insurance premium)
Insurance covers all damage, so expected out-of-pocket = $400
Without insurance: Expected cost = $300
With insurance: Expected cost = $400
Since $300 < $400, insurance is not recommended at 3% probability
E[Loss without insurance] = $10,000 × 0.08 + $0 × 0.92 = $800
E[Cost with insurance] = $400 (same premium)
Since $800 > $400, insurance is recommended at 8% probability
Break-even when: Probability × $10,000 = $400
Probability = $400/$10,000 = 0.04 = 4%
At 3% probability, don't buy insurance (expected cost $300 vs $400).
At 8% probability, buy insurance (expected cost $800 vs $400).
Break-even occurs at 4% probability.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Insurance Decision: Buy if E[Loss] > Premium
• Break-even Analysis: Set E[Cost without] = E[Cost with]
Total Expected Cost: Purchase price + Expected repair cost.
Repair Probability: Likelihood of requiring repairs within specified time frame.
E[Repair Cost A] = $200 × 0.10 + $0 × 0.90 = $20
Total Cost A = Purchase Price + E[Repair Cost]
= $800 + $20 = $820
E[Repair Cost B] = $150 × 0.05 + $0 × 0.95 = $7.50
Total Cost B = Purchase Price + E[Repair Cost]
= $1000 + $7.50 = $1007.50
Total Cost A = $820, Total Cost B = $1007.50
Since $820 < $1007.50, Phone A has the lower expected total cost
Phone A: Lower upfront cost but higher repair probability
Phone B: Higher upfront cost but lower repair probability
Phone A has the lower expected total cost of $820 compared to Phone B's $1007.50.
Based on expected value, Phone A is the better financial choice.
• Expected Value Formula: E[X] = Σ [x × P(X = x)]
• Total Expected Cost: Purchase Price + E[Future Costs]
• Decision Rule: Choose option with lowest expected total cost
Complete Summary: Probability Distributions Applications in Games and Decisions
Expected Value: The weighted average of all possible outcomes, weighted by their probabilities.
Net Value: Gross outcomes minus associated costs.
Decision Rule: Choose option with highest expected value (for risk-neutral decision makers).
Game Fairness: E[Net] = 0 (fair), E[Net] > 0 (favorable), E[Net] < 0 (unfavorable).
Linearity: E[aX + b] = aE[X] + b for constants a, b.
Additivity: E[X + Y] = E[X] + E[Y] for any two random variables.
Probability Weighting: More likely outcomes contribute more to expected value.
Long-run Average: The theoretical average over infinite repetitions.
Step 1: Identify all possible outcomes and their probabilities.
Step 2: Calculate net value for each outcome.
Step 3: Apply expected value formula: E[X] = Σ [x × P(X = x)].
Step 4: Compare options and make decision based on expected values.
Always Calculate Net: Subtract costs from gross amounts to get net outcomes.
Verify Probabilities: Ensure all probabilities sum to 1.
Consider Context: Expected value is optimal for risk-neutral decision makers.
Look for Break-even Points: Find thresholds where decisions change.
Expected Value: The theoretical average value of a random variable over many repetitions of an experiment. It represents the center of the probability distribution.
Risk-Neutral Decision Maker: Someone who makes decisions based solely on expected value, ignoring the variance or risk involved.
Utility Theory: A framework that accounts for risk preferences beyond just expected value, recognizing that people may value certainty differently.
Opportunity Cost: The value of the next best alternative foregone when making a decision.
Present Value: The current value of a future payment or stream of payments, discounted at an appropriate rate.
- Define the Problem: What decision needs to be made?
- Identify Alternatives: List all possible choices.
- Determine Outcomes: For each alternative, identify all possible outcomes.
- Assign Probabilities: Estimate probability of each outcome.
- Calculate Values: Determine monetary or utility value of each outcome.
- Compute Expected Values: Apply E[X] = Σ [x × P(X = x)] for each alternative.
- Make Decision: Choose alternative with highest expected value.
• Expected Value: E[X] = Σ [x × P(X = x)]
• Linear Transformation: E[aX + b] = aE[X] + b
• Additivity: E[X + Y] = E[X] + E[Y]
• Net Calculation: Net = Gross - Costs
Questions & Answers
Question: In Exercise 1, the expected value was positive, meaning the game is favorable to the player. But shouldn't casinos always have the edge? How is this possible?
Answer: You're correct that real casinos always have a positive edge! This is just a hypothetical example to demonstrate the concept.
Real Casino Games:
- Blackjack house edge: 0.5-2% (with basic strategy)
- Roulette house edge: 2.7% (European) or 5.26% (American)
- Slot machines: 2-15% house edge
Why Casinos Must Have Edge:
- They need to cover operational costs (staff, rent, maintenance)
- They need to make a profit to stay in business
- They must pay taxes and dividends
In our example, a real casino would adjust the payouts to ensure a positive house edge, making E[Net] negative for the player.
Question: In Exercise 4, you said not to buy insurance at 3% probability. But wouldn't someone want insurance anyway to protect against the catastrophic loss of $10,000?
Answer: Excellent observation! You've identified a key limitation of pure expected value analysis.
Expected Value vs. Risk Management:
- Expected value is optimal for risk-neutral individuals
- Most people are risk-averse and willing to pay extra to avoid large losses
- Insurance provides peace of mind and protection against catastrophic losses
Utility Theory Perspective:
The utility (satisfaction) of avoiding a $10,000 loss may exceed the disutility of paying $400, even though E[Cost] is higher. Many people prefer the certainty of paying $400 over the 3% chance of losing $10,000.
Expected value is a starting point for decision analysis, but real decisions often incorporate risk tolerance and personal values.
Question: Most lotteries have very low odds of winning but huge prizes. Why do people still buy lottery tickets if the expected value is so negative?
Answer: This is a fascinating behavioral economics question! People don't always make decisions based on expected value:
Psychological Factors:
- Prospect Theory: People overweight small probabilities and underweight large ones
- Utility of Large Wins: The utility of $1 million isn't proportional to the utility of $1
- Dream Factor: The entertainment value of imagining winning
Behavioral Economics:
People often focus on the possibility of winning rather than the probability. The excitement and hope of winning can provide utility that exceeds the cost of the ticket.
Additionally, for some people, the cost of a ticket is so small that the negative expected value is acceptable for the chance of a life-changing prize.