Simple Functions
GLOSSARY/ANALYSIS

Node Probability

Node probability is the estimated likelihood assigned to a specific node in a causal tree. Changes to leaf node probabilities cascade upward through the tree, updating the root probability and all derived edge calculations.

CLI:sf thesis view --tree

How Node Probabilities Work

Every node in a causal tree has a probability — the estimated chance that the condition described by that node is or will be true.

Leaf Nodes vs. Branch Nodes

Leaf nodes have probabilities you assign directly based on evidence:

  • "Unemployment exceeds 5%" — you estimate 35% based on current trends

Branch nodes derive their probabilities from their children:

  • If children are connected by AND logic: branch prob = child1 × child2
  • If children are connected by OR logic: branch prob = 1 - (1-child1) × (1-child2)
  • Many branches use weighted combinations

Updating Node Probabilities

When new information arrives, the agent maps it to the relevant node and adjusts the probability. This update propagates:

  1. New data: February unemployment rate comes in at 4.8% (up from 4.5%)
  2. Maps to: node n1.1 (unemployment > 5%)
  3. Update: n1.1 probability increases from 35% to 42%
  4. Propagation: n1 (labor market deterioration) increases
  5. Root update: Overall recession probability increases
  6. Edge recalculation: Edge on KXRECSSNBER-26 widens

The Power of Granular Updates

Without a causal tree, you'd have to answer: "How does a 0.3% increase in unemployment affect the overall probability of recession?" That's a hard, fuzzy question.

With a causal tree, you answer: "How does this data affect the probability of unemployment exceeding 5%?" That's much easier — and the math handles the rest.

Example

Before February unemployment data:
  n1.1: Unemployment > 5%     p = 35%
  n1:   Labor deterioration    p = 50%
  Root: Recession              p = 38%
  Edge on KXRECSSNBER-26:      38% - 28% = +10pt

After data (unemployment = 4.8%, up from 4.5%):
  n1.1: Unemployment > 5%     p = 42%  (+7pt)
  n1:   Labor deterioration    p = 55%  (+5pt)
  Root: Recession              p = 42%  (+4pt)
  Edge on KXRECSSNBER-26:      42% - 29% = +13pt

  (Market moved from 28¢ to 29¢, but thesis moved more)

Related Terms

Thesis Confidence Score

Confidence is a 0-100% score that measures how certain you are about your probability estimates, not the probability itself. High confidence means your estimates are well-researched; low confidence means they're speculative.

Causal Tree

A causal tree is a structured probabilistic model that breaks down a complex event into a hierarchy of independent, verifiable sub-conditions. Each node has a probability, and the tree computes the overall probability of the root event.

Thesis

A thesis is a structured investment argument that combines a directional view, a causal model of why the view is correct, and specific market positions that express the view. In SimpleFunctions, a thesis is the core unit of analysis.

Evaluation Cycle

An evaluation cycle is the automated process by which the agent reviews new signals, updates causal tree probabilities, recalculates edge, and determines whether any strategy conditions are met.

← All glossary termsRead the docs →