Decision Tree & Payoff Calculator

A render engine that turns structured decision data into visual + numerical decision-support output. It takes as input a list of actors, each actor's moves, branch probabilities, and 4-dimensional (economic / emotional / reputational / relational) payoff vectors; it produces a mermaid graph TD decision tree, a side-by-side payoff matrix table for each main branch, and an expected value (EV) calculation for each move.

When to use it

• When making a quantitative comparison of multi-move decision situations such as negotiations, raise requests, partnership splits, or price haggling.
• When you need to visualize for the user the branches produced by a strategy / game-theory / change-management agent.
• When you want to show the high variance of counter-intuitive ⚡ moves in concrete numbers.

What it produces

• input.json — the root decision, the EV weights of the 4 dimensions (summing to 1.0), and each branch's probability + payoff vector.
• output.json — the weighted payoff for each branch, its EV contribution, the total EV per main branch, and finally the EV ranking (top 1-3).

It supports 3 behavior layers: rational, bounded, bias-triggered—each distinguished by a different visual style. Branches missing a probability or payoff are skipped with a warning; if no weights are given, 0.25/0.25/0.25/0.25 is assumed. The output is a generic, reusable component that can be shown directly to the user and reused by any decision/strategy agent.