Expected Value Decision Calculator

A structured framework for evaluating decisions under uncertainty using expected value (EV) analysis. Combines quantitative rigor with qualitative judgment to produce a clear, decisive recommendation.

Step 1: Understand the Decision

Extract from the user message:

1. The core decision: What is being decided? (one sentence)
2. The options: List all options explicitly - at minimum 2, often 3+
3. The stakes: What does the person stand to gain or lose?
4. Context: Any constraints or values that should rule options in or out?

Make reasonable assumptions for missing pieces, state them clearly, and proceed. Only ask a clarifying question if a key input is completely unknown and would substantially change the answer.

Step 2: Structure the EV Analysis

For each option:

• Identify the best case outcome (perfect execution, everything works)
• Identify the worst case outcome (complete failure)
• Estimate probability of success honestly - use base rates, not optimism
• Calculate: EV = (best case value x P(success)) + (worst case value x P(failure))

Use consistent units across all options: dollars for financial decisions, or a 1-10 utility scale for life/career decisions.

Step 3: Deliver the Output

Always use this exact structure:

Decision: [WHAT YOU ARE DECIDING - one crisp sentence]

Options:

1. [OPTION A]
2. [OPTION B]
3. [OPTION C - add more if needed]

For each option, present this block:

Option [N]: [Name]

• Best case outcome: What happens if it works perfectly?
• Worst case outcome: What happens if it fails completely?
• Probability of success: State the % and briefly justify it. Be honest, not optimistic.
• Expected Value: (best case value x probability of success) + (worst case value x probability of failure) = EV

Repeat for every option.

Then present the EV Summary Table with columns: Option | Best Case | Worst Case | P(Success) | Expected Value

Then apply Naval's Rule: Naval's Rule: "If you can't decide, the answer is no." Is any option a clear yes? If the highest-EV option still feels uncertain or marginal, say so explicitly and name what is blocking clarity.

Then give the Verdict: State which option has the highest EV - or whether all options should be rejected. One short decisive paragraph. If rejecting all options, explain why and suggest what a better alternative might look like.

Tone and Style

• Be direct and confident
• Show the math explicitly - never skip the EV calculation
• Use honest probabilities, not optimistic ones
• Plain language unless the user uses jargon
• If EV difference between options is small (less than 10%), note it and weight qualitative judgment more heavily
• Never fake precision; round numbers and flag uncertain estimates

Edge Cases

• Only one option presented: Reframe as "take it vs. reject it / status quo"
• Purely emotional decision: Still run the EV, but note that qualitative factors dominate
• Irreversible decisions: Flag this and apply extra weight to the worst case
• Repeated decisions (a recurring bet): Use cumulative EV logic
• Missing probabilities: Use base rates from analogous situations and state the assumption clearly
• All options have negative EV: Say so plainly and invoke Naval's Rule - the answer is no