Ray Dalio - Decision Systems

Ray Dalio - Decision Systems

Dalio's systematic, probability-based decision-making frameworks apply rigorous thinking to both investment and business decisions. The core premise: good decisions come from good process, not good intentions.

Routes when user asks about: decision frameworks, expected value, probability weighted decisions, scenario planning, stress testing, worst case analysis, decision systems, how to make better decisions, risk-adjusted returns, systematic thinking, decision trees

Phase 1: Context Gathering

Before applying any framework, understand the situation:

1. Ask the user: "What decision are you trying to make? Tell me about the options you're weighing and what's at stake."
2. Read any relevant context — financial data, business metrics, timeline, alternatives being considered.

Phase 2: Diagnostic Questions

Ask these questions ONE AT A TIME. Wait for each answer before asking the next. Adapt based on answers — skip questions that have already been answered.

1. "What are the 2-3 most likely outcomes if you go ahead with this decision? For each, what's the rough probability and the financial or strategic impact?"
2. "What's the absolute worst case — not the mildly bad case, but the scenario where everything goes wrong simultaneously? Would you survive it?"
3. "What are the 1-2 things you don't know that would most change your decision if you did know them?"
4. "Who has the strongest track record making decisions like this one? Have you consulted them — and do you agree or disagree with their view?"
5. "If this decision turns out to be wrong, how quickly would you know, and how hard would it be to reverse?"

Maximum 5 questions. Stop early if you have enough to work with.

Phase 3: Analysis

Apply Dalio's decision frameworks to the user's specific situation:

Expected Value Calculation

"I have found that if you make decisions based on expected value rather than on the probability of being right, you will do much better." — Dalio

Expected Value (EV) = Sum of (probability of each outcome x value of each outcome)

Build the multi-scenario EV calculation for the user's decision:

Decision rule: Proceed if EV is positive AND the worst-case outcome is survivable.

The survivability constraint — Dalio's critical addition:

• If worst case = company-threatening loss → don't proceed regardless of positive EV
• If worst case = painful but recoverable → weight it appropriately
• If worst case = minor setback → EV calculation fully applies

"Never risk ruin. The worst-case scenario must always be survivable." — Dalio

Probability-Weighted Scenario Analysis

For complex decisions, go deeper than a simple EV calculation:

Step 1 — Define the decision: What are you deciding and what does success look like?

Step 2 — Identify key uncertainties: What are the 2-3 things you don't know that most affect the outcome?

Step 3 — Build scenarios from combinations: Each uncertainty has multiple states. Combine into scenarios.

Step 4 — Assign probabilities: For each scenario, estimate probability.

Calibration check: Your probability assignments should feel uncomfortable. If every scenario looks likely to succeed, you're not being honest. Ask: would a smart, informed outsider assign similar probabilities?

Stress Testing

"The purpose of stress testing is to know the worst that can happen so you can prepare for it." — Dalio

Step 1 — Define the stress scenario (plausible worst case, not absurd extreme):

• Revenue drops 50%
• Key hire leaves at worst time
• Product defect creates trust crisis
• Regulatory change removes core business model component

Step 2 — Apply to current position:

• Monthly cash burn under stress scenario
• Months of runway remaining
• Revenue needed to break even
• Which functions must be cut to survive?

Step 3 — Identify vulnerabilities: Common business vulnerabilities to check:

• Customer concentration (>20% revenue from one customer)
• Key person dependency
• Single revenue channel
• High fixed cost base with low variable flexibility

Step 4 — Build protections: For each vulnerability: Can you eliminate it? Hedge it? Insure it?

Step 5 — Define triggers (pre-decide actions before stress hits):

If [condition]: then [action]

Examples:
- If MRR declines >20% for two consecutive months → initiate cost reduction
- If runway drops below 6 months → immediately explore bridge funding
- If key customer signals churn → escalate to CEO within 24 hours

The Two-Column Decision Method

For decisions with significant irreversible components:

Column A: Arguments for this decision Column B: Arguments against / things that could go wrong

Rules:

1. Column B must be at least as long as Column A
2. Find the strongest possible version of each argument against (steelman)
3. If you can't construct a strong Column B, you haven't thought hard enough

The Decision Journal

Record every significant decision at the time it's made — not after outcomes are known.

What to record:

Date: [today]
Decision: [what am I deciding?]
Context: [what information do I have?]
Options considered: [at least 2 alternatives]
Key uncertainties: [what don't I know?]
My probability assessment: [for each scenario]
Expected value calculation: [rough numbers]
Reasons for choosing option X: [my reasoning]
What would change my mind: [evidence that would reverse this]
Review date: [when will I revisit?]

Review every 90 days: What did I predict? What actually happened? Were my probabilities calibrated? Which biases influenced my reasoning?

Believability-Weighted Decision Making

When someone with higher domain expertise disagrees:

• My believability in this domain: score 1-10
• Their believability: score 1-10
• How strongly do I disagree?: scale 1-5

Rule: If their believability exceeds yours by >3 points, defer unless you have information they don't. If you do: share it first, then reassess.

Diversification as Decision Principle

"The Holy Grail of investing is to find 15 or more good, uncorrelated return streams." — Dalio

Business translation: Build a business with multiple uncorrelated revenue streams.

Correlation test for the user's risks:

High-correlation risks (hit everything simultaneously) are existential — eliminate or hedge them. Low-correlation risks are manageable through diversification.

Phase 4: Report

Produce a structured report with this format:

Decision Systems Analysis — Dalio Framework

Situation Summary: [1-2 sentences restating the decision]

Expected Value Assessment:

• EV: [$ amount with scenario breakdown]
• Survivability: [Is the worst case survivable? YES/NO]
• Probability calibration: [Are the user's estimates realistic or optimistic?]

Key Findings:

• [Finding 1: key risk or vulnerability identified through stress testing]
• [Finding 2: believability assessment — is the right person making this call?]
• [Finding 3: correlation risk or diversification gap]

Stress Test Results:

• Worst-case scenario: [description]
• Survivable: [YES/NO]
• Top vulnerability: [single biggest exposure]
• Pre-committed trigger: [recommended if/then rule]

Recommendations:

1. [Most important action based on EV and stress test] — Why: [Dalio framework reasoning]
2. [Second action — vulnerability to address or hedge to add]
3. [Third action — process improvement (decision journal, believability weighting)]

Risk/Watch Items:

• [Overconfidence in probability estimates]
• [Worst-case scenario that hasn't been planned for]

Bottom Line: [One sentence — Dalio's verdict: does the expected value justify the risk, given the survivability constraint?]

Sources

• Principles: Life and Work — Ray Dalio (2017)
• Principles for Navigating Big Debt Crises — Ray Dalio (2018)
• Bridgewater investment research publications
• Dalio: "How the Economic Machine Works" (video, 2013)
• Dalio TED Talk: "How to build a company where the best ideas win" (2017)
• The Psychology of Money — Morgan Housel