Reasoning Frameworks
A field guide to structured thinking. Learn how to choose the right reasoning pattern for explanation, judgment, decision-making, diagnosis, and strategy.
12 categories • 92 reasoning frameworks
What These Frameworks Are For
Reasoning frameworks are reusable structures for thinking. They help you avoid relying on mood, habit, or rhetorical confidence alone by giving you a more explicit way to move from inputs to conclusions.
Some frameworks are built for certainty, like deduction. Others are built for uncertainty, like Bayesian updating or scenario analysis. Some help you explain and defend a claim. Others help you diagnose causes, choose among options, or think clearly inside complex systems.
The goal is not to memorize labels. The goal is to develop a stronger instinct for matching the thinking tool to the problem in front of you.
How To Use This Library
Start with the kind of problem you are facing: explanation, prediction, choice, diagnosis, persuasion, or coordination.
Choose one framework that fits the situation instead of stacking many at once.
Use the framework to surface assumptions, not to pretend you have more certainty than you do.
After reaching a conclusion, test it with a second framework that looks for different failure modes.
A Simple Starting Path
If you need a conclusion
Begin with deductive, inductive, or abductive reasoning depending on whether your evidence supports certainty, probability, or best explanation.
If you need a decision
Start with expected value, scenario analysis, sensitivity analysis, or value of information to separate choice quality from outcome luck.
If you need a diagnosis
Reach for causal inference, root cause analysis, Five Whys, or fault trees when the real task is understanding what is driving the outcome.
Core Inference
Use these when you need to move from evidence to conclusion with more discipline, whether you are generalizing, explaining, comparing, or testing alternatives.
7 frameworks
Deductive Reasoning
Derive logically certain conclusions from premises (if premises true and form valid).
Inductive Reasoning
Generalize from observations to probabilistic conclusions.
Abductive Reasoning
Infer the best explanation for observed facts.
Analogical Reasoning
Transfer structure from a known domain to a new one based on similarity.
Counterfactual Reasoning
Reason about 'what would have happened if…' to test causality and decisions.
Case-Based Reasoning
Solve new problems by adapting solutions from similar past cases.
Default Reasoning
Assume typical rules apply unless evidence indicates exceptions.
Uncertainty & Probability
Use these when the world is noisy, evidence is incomplete, and your job is not certainty but better calibrated judgment.
7 frameworks
Bayesian Updating
Update beliefs by combining priors with new evidence.
Frequentist Statistical Inference
Estimate parameters and test hypotheses using sampling distributions and p-values.
Likelihood Thinking
Compare hypotheses by how well they predict the data (likelihood ratios).
Base Rate Reasoning
Anchor judgments in population rates before using case-specific evidence.
Expected Value Reasoning
Evaluate choices by probability-weighted outcomes.
Risk vs Uncertainty Separation
Distinguish quantifiable risk from unquantifiable uncertainty (Knightian).
Calibration Practice
Train probabilistic judgment by scoring forecasts against outcomes.
Decision Analysis
Use these when you are choosing between options, weighing tradeoffs, or stress-testing a plan across different futures.
13 frameworks
Expected Utility Theory
Choose actions that maximize expected utility rather than expected value.
Sensitivity Analysis
Test how conclusions change with assumptions and parameter variation.
Scenario Analysis
Evaluate decisions across multiple plausible futures.
Monte Carlo Simulation
Estimate outcomes by simulating many random draws from uncertain variables.
Value of Information
Quantify whether gathering more information is worth the cost/time.
Decision Tree Analysis
Model sequential decisions with probabilities and payoffs.
Influence Diagrams
Compact causal/decision graph connecting choices, uncertainties, and objectives.
Multi-Criteria Decision Analysis (MCDA)
Compare options across weighted criteria.
Analytic Hierarchy Process (AHP)
Pairwise comparisons to derive weights and rank alternatives.
Cost–Benefit Analysis
Quantify costs and benefits to guide choices.
Regret Minimization
Choose actions that minimize expected regret or worst-case regret.
Premortem
Assume failure happened; generate plausible causes to mitigate upfront.
Postmortem / After-Action Review
Extract lessons from outcomes and update processes.
Scientific Reasoning
Use these when you want explanations that can be tested, challenged, replicated, and improved rather than merely asserted, including checking whether your model still fits reality.
9 frameworks
Scientific Method
Hypothesize, test, measure, revise; emphasize reproducibility and falsifiability.
Hypothetico-Deductive Method
Derive testable predictions from hypotheses and attempt to refute them.
Falsificationism
Prefer theories that survive serious attempts at refutation.
Inference to the Best Explanation (IBE)
Select hypotheses that best explain the evidence with parsimony and fit.
Operationalization
Define measurable proxies for abstract concepts to test claims.
Model Selection
Choose among models using predictive accuracy and complexity penalties.
Replication & Triangulation
Validate findings via repeated studies and independent methods.
All Models Are Wrong (But Some Useful)
Treat models as tools with domains and failure modes, not truth.
Map–Territory Check
Explicitly list what your representation ignores and where it might fail.
Causality
Use these when you need to separate correlation from cause, diagnose failure, or understand what would actually change an outcome.
8 frameworks
Causal Inference with DAGs
Use causal graphs to reason about confounding, interventions, and identification.
Counterfactual Causal Models
Model interventions and counterfactuals (do-operator style thinking).
Randomized Controlled Trials (RCTs)
Infer causal effects via random assignment.
Quasi-Experimental Designs
Infer causality when randomization isn’t possible (DiD, RDD, IV, matching).
Root Cause Analysis (RCA)
Systematically trace from symptoms to underlying causes.
Five Whys
Iteratively ask 'why' to dig from symptom to cause.
Fishbone / Ishikawa Diagram
Categorize possible causes to structure investigation.
Fault Tree Analysis
Top-down logical decomposition of how failures can occur.
Argumentation
Use these when you need to present reasoning clearly, support claims with evidence, and make disagreements easier to evaluate.
8 frameworks
Toulmin Argument Model
Claim–Data–Warrant–Backing–Qualifier–Rebuttal structure.
Claim–Evidence–Reasoning (CER)
A simple framework to build evidence-backed explanations.
Pragma-Dialectical Argumentation
Rules for critical discussion aimed at resolving differences reasonably.
Argument Mapping
Visualize premises, objections, and support to assess structure and gaps.
Steelmanning
Reconstruct the strongest version of an opposing argument before evaluating it.
Principle of Charity
Interpret others’ statements in the most reasonable way consistent with evidence.
Burden of Proof
Assign responsibility for providing evidence based on claim strength and context.
Fallacy Checking
Detect invalid forms and misleading rhetoric (informal and formal fallacies).
Problem Structuring
Use these when a problem is too tangled to attack directly and needs to be broken into clear, addressable pieces first.
7 frameworks
MECE (Mutually Exclusive, Collectively Exhaustive)
Partition a problem space cleanly without overlaps or gaps.
Issue Tree / Logic Tree
Decompose a question into sub-questions until actionable.
Hypothesis-Driven Development
Start with a testable hypothesis, then seek evidence to confirm/refute.
Problem Framing & Reframing
Test alternative problem statements to uncover better solution spaces.
Constraint-Led Thinking
Use constraints to sharpen choices and reveal trade-offs.
First Principles Decomposition
Break assumptions down to fundamentals and rebuild from ground truth.
Inversion
Solve by asking how to cause failure or the opposite outcome, then avoid it.
Systems & Operational Reasoning
Use these when outcomes emerge from feedback loops and interactions, and thinking needs to turn into disciplined, repeatable execution.
10 frameworks
Systems Thinking
Reason about interconnected components, feedback, and emergent behavior.
Complex Adaptive Systems
Model agents adapting locally leading to emergent macro-patterns.
Cynefin Framework
Classify contexts (clear/complicated/complex/chaotic) to choose response style.
Ashby’s Law of Requisite Variety (Intuitive)
Controllers need sufficient variety to regulate system variety.
Bottleneck / Theory of Constraints
Improve system throughput by addressing the limiting constraint.
OODA Loop
Observe–Orient–Decide–Act cycles for fast adaptation.
PDCA Cycle
Plan–Do–Check–Act for continuous improvement.
DMAIC (Six Sigma)
Define–Measure–Analyze–Improve–Control for process improvement.
After-Action Review (AAR)
Structured reflection: expected vs happened, why, and what to change.
Double-Loop Learning
Improve not just actions but underlying assumptions and governing variables.
Dialogue & Sensemaking
Use these when thinking needs to be externalized and worked through, alone or with others, to become clearer.
5 frameworks
Socratic Method
Use probing questions to expose assumptions and refine beliefs.
Ladder of Inference
Track how observations become interpretations and beliefs to avoid leaps.
Sensemaking Loop
Iterate between data, frames, narratives, and action to reduce uncertainty.
Rubber Duck Debugging (Articulation)
Explain aloud to uncover gaps and contradictions.
Six Thinking Hats
Force perspective shifts (facts, emotions, risks, benefits, creativity, process).
Creative Reasoning
Use these when you are generating possibilities, reframing a problem, or escaping the gravity of first-obvious answers.
5 frameworks
Design Thinking
Empathize–Define–Ideate–Prototype–Test for human-centered solutions.
TRIZ
Inventive problem-solving via contradiction analysis and solution patterns.
SCAMPER
Idea generation via Substitute/Combine/Adapt/Modify/Put to use/Eliminate/Reverse.
Morphological Analysis
Explore combinations of parameter values to generate solution spaces.
Divergent → Convergent Cycle
Expand options broadly, then narrow via evaluation and constraints.
Group & Adversarial Reasoning
Use these when judgment is formed collectively, or when claims may be misleading and someone may be trying to exploit your blind spots.
8 frameworks
Red Teaming
Actively attack plans/models to uncover hidden vulnerabilities.
Devil’s Advocate
Assign a role to challenge assumptions and surface counterarguments.
Threat Modeling (Generalized)
Identify assets, adversaries, attack paths, and mitigations.
Precommitment
Bind future actions to resist predictable failures or temptations.
Delphi Method
Iterative expert elicitation to converge on better forecasts/estimates.
Wisdom of Crowds (Aggregation)
Aggregate diverse independent judgments to improve accuracy.
Prediction Markets (Conceptual)
Use market prices as aggregated probabilistic forecasts.
Nominal Group Technique
Structured idea generation and voting to reduce groupthink.
Strategy
Use these when other people's incentives, likely responses, and competitive constraints materially shape the result.
5 frameworks
SWOT Analysis
Strengths/Weaknesses/Opportunities/Threats to structure strategic thinking.
Porter’s Five Forces
Analyze industry structure via competitive forces.
Jobs to Be Done (JTBD)
Frame demand as progress customers hire products to achieve.
Theory of Change
Map how interventions lead to outcomes via assumptions and causal pathways.
Balanced Scorecard
Reason across multiple performance perspectives (financial, customer, process, learning).