Home/Blog/Expected Value: The Single Most Useful Concept in Decision-Making
Applied Probability7 min read

Expected Value: The Single Most Useful Concept in Decision-Making

Expected value is the average outcome of a random process over many repetitions. It is arguably the single most useful concept in decision-making under uncertainty.

P
The Probability Lab Team
July 2, 2025

Expected value (EV) is the average outcome of a random process over many repetitions. It is calculated by multiplying each possible outcome by its probability and summing the results. It is, arguably, the single most useful concept in decision-making under uncertainty — and one of the most systematically ignored by human intuition.

The basic calculation

A simple example: you are offered a bet. Roll a six-sided die. If it shows 6, you win $10. Otherwise you lose $1.

Simple EV Calculation
EV = (1/6 × $10) + (5/6 × −$1)
EV = $1.667 − $0.833
EV = +$0.833 per play

Over 100 plays: expected gain ≈ $83

When EV says one thing and instinct says another

Consider: flip a coin. If heads, you win $110. If tails, you lose $100. EV = (0.5 × $110) + (0.5 × -$100) = +$5. Most people refuse this bet. Loss aversion makes the potential $100 loss feel worse than the potential $110 gain feels good. Kahneman and Tversky documented this systematically in Prospect Theory.

The Kelly Criterion: sizing bets by EV

Knowing a bet has positive EV is not enough. You also need to know how much to bet. The Kelly Criterion solves this:

Kelly Criterion
f* = (bp − q) / b

Where:
  f* = fraction of bankroll to bet
  b  = net odds (win amount / stake)
  p  = probability of winning
  q  = 1 − p

For the coin flip (win $110, lose $100, p = 0.5):
  f* = (1.1 × 0.5 − 0.5) / 1.1 = 0.0455 = 4.55%

Optimal bet: 4.55% of your bankroll per flip.

Kelly betting is used by professional gamblers, quantitative traders, and venture capital portfolio managers. It maximizes long-run bankroll growth while avoiding ruin.

EV in non-gambling decisions

DecisionEV frameworkCaveat
Buying insuranceEV is negative (insurer must profit)Catastrophic loss avoidance has non-financial value
Taking an umbrellaP(rain) × cost of wet vs. P(no rain) × inconvenienceSimple, works well
Accepting a job offerSalary × career growth potential minus stress costsNon-quantifiable factors matter too

The limitation of EV

EV is the right framework for decisions made repeatedly in a stable environment with bounded stakes. It is a less complete framework for one-time decisions with catastrophic downside, decisions involving deeply personal utility, and decisions where probability estimates are highly uncertain.

The solution in most cases is not to abandon EV but to weight it alongside variance — prefer lower variance for the same EV when stakes are high relative to your wealth.

Our Expected Value Calculator lets you add any number of outcomes with probabilities and payoffs and calculates the EV in real time. It also validates that your probabilities sum to 100% — a basic but frequently violated constraint in informal decision-making.