Odds describe the situation, decisions use them
Knowing that a die has a one-in-six chance of showing a particular face is a fact about the world. It does not tell you what to do. A decision needs one more ingredient: what you gain or lose in each outcome. Put chances and payoffs together and you can compare options on equal terms. The standard tool for that is expected value.
Expected value is the average result you would get if the same choice could be repeated a very large number of times. It is not a prediction for the next attempt. On any single play you receive one specific outcome, never the average itself. What the average gives you is a fair way to rank choices when you cannot know how a single one will turn out.
How to calculate it
The method has three steps.
- List every possible outcome of the choice.
- Multiply each outcome’s value by its probability.
- Add the results.
Here is a deliberately simple illustration. Suppose a game costs 2 points to enter. You roll one fair die and receive 6 points if it shows a six, otherwise nothing.
- Chance of a six: 1 in 6.
- Average payout: 6 x 1/6 = 1 point.
- Cost to play: 2 points.
- Expected value of playing: 1 - 2 = -1 point per play.
On average you lose one point each time. You might still win on the first roll, but across many rolls the average is what you should expect to see. Compare that with a different invented game: you pay 1 point, flip a fair coin, and receive 3 points on heads. The average payout is 3 x 1/2 = 1.5, so the expected value is 1.5 - 1 = +0.5 per play. Games like this are useful for practice, but real gambling products do not offer favourable terms like this one.
Comparing options, not judging one in isolation
The method becomes more useful when you have two or more real choices. Imagine a board game where you can take a safe action worth 10 points, or a risky action that gives 30 points with probability 0.4 and nothing otherwise. The risky option averages 30 x 0.4 = 12 points, so it is ahead of the safe 10 on average. Whether to take it still depends on the situation, which leads to the main caveats.
A second comparison: likely small gain versus unlikely big one
Options often differ in shape as well as size. Suppose, again as an invented illustration, that you can choose between two moves in a card game.
- Move 1: an 80 percent chance to gain 5 points and a 20 percent chance to lose 5 points. Expected value: 0.8 x 5 + 0.2 x (-5) = 4 - 1 = +3.
- Move 2: a 10 percent chance to gain 40 points and a 90 percent chance to lose 2 points. Expected value: 0.1 x 40 + 0.9 x (-2) = 4 - 1.8 = +2.2.
Move 2 has the more exciting top prize, but Move 1 is ahead on average, 3 against 2.2. It is also steadier: Move 2 loses nine times out of ten. The calculation does not force you to choose Move 1, since you may value a shot at the big prize, but it shows you exactly what that preference costs on average.
What expected value leaves out
Expected value is a starting point rather than a verdict. Several things sit outside the calculation.
Spread of outcomes. Two options can share the same average and feel very different. A guaranteed 10 and a coin flip for 20 or nothing both average 10. The difference between them is described by volatility and return, and which you prefer depends on what you can afford to lose and what the game is asking of you.
Stakes you cannot replace. If a loss would end your participation, such as in a knockout round, the average matters less than the chance of surviving. The idea of risk of ruin deals with exactly this situation.
Probabilities you only estimate. In a card game or a strategy game you rarely know the true odds. Your estimate may be rough, and a rough input produces a rough answer. A decision based on a careful estimate is still better than one based on a hunch, but it is not certain.
Value that is not points. Enjoyment, learning and time all count for something in a game played for fun. The calculation only covers what you put numbers on.
Good decisions and good outcomes are different things
A choice with positive expected value can lose, and a poor choice can win. Because results are noisy, you cannot judge a decision by a single outcome. The law of large numbers says that averages settle down as trials accumulate, but it says nothing about the next trial and it does not undo earlier ones. Past results do not change future odds: if a fair coin has landed heads five times, the next flip is still 50-50.
This is why informed play focuses on process. Ask whether the decision was sound given what you knew, and keep a record so that you can look for patterns over a longer run.
Where games with a house edge fit
In casino-style games the payouts are set so that the average result for the player is negative. That built-in margin is known as the house edge, and the same expected-value arithmetic reveals it. When every bet carries negative expectation, no choice of bet turns the game into a profitable one over time. Playing carefully can reduce how fast you lose, for example by choosing bets with a smaller margin, but it cannot reverse the direction. If you play such games, treat the cost as entertainment spending, set it in advance, and do not borrow or chase losses. Gambling can be harmful, and anyone who feels it is becoming a problem should contact a local support service.
Where the method is most useful
Expected value shines in games where decisions matter and the odds can be worked out or reasonably estimated: tabletop and card games, puzzle-like choices, and strategy decisions where you are weighing a risky move against a safe one. In poker the same idea shows up as pot odds, which compare the price of a call with the chance of completing a hand. A single formula is not the whole of good judgement, but it forces you to write down the chances and the payoffs, and that alone prevents many errors.
A short checklist for any decision
- What are all the outcomes, including the ones that feel unlikely?
- How likely is each, and how sure am I of that number?
- What is each outcome worth to me, in the same units?
- What is the average, and what is the worst case?
- Can I afford the worst case?
Answering these takes a minute and gives you a decision you can explain afterwards, whether it works out or not.