All demos
Expected-value model
Turn probabilities and payoffs into one comparable number, then watch what one wrong guess does to it.
+10 pts here +0.30
+10 pts here −0.15
+10 pts here −0.79
Expected value5.7
Weights add up to 100%
The last column re-runs the model with 10 points moved onto each outcome in turn: if the number barely moves, the decision is robust — if it flips, it is a coin toss.
How it works
Each outcome contributes probability × payoff, and the sum is the expected value. It is a decision aid, not a prediction. The last column moves 10 points onto each outcome in turn and recomputes the total, which is usually enough to see whether the answer is robust or a coin toss. Weights are normalised first, so a table that does not add up to 100 still gives a comparable number.
Built withTypeScript · AI: decision models (JEV)