A library for thinking under uncertainty.
Cognitive biases, tail risks, antifragility, decision protocols. Vladyslav's practical notes on Kahneman and Taleb - turned into a searchable system for better decisions.
Concepts and mental models
System 1 & System 2
Two modes of thinking: System 1 is fast, automatic, associative; System 2 is slow, deliberate, effortful.
Open conceptWYSIATI - What you see is all there is
We build confident conclusions from the small slice of information available, ignoring what we don't know.
Open conceptCognitive ease
When information feels easy to process - familiar, clear, repeated - we trust it more, even when it's wrong.
Open conceptQuestion substitution
When a question is hard, the mind silently answers an easier related question instead - and reports the answer as if it answered the original.
Open conceptAvailability heuristic
We judge how common or likely something is by how easily examples come to mind.
Open conceptAnchoring
Any number you see first quietly pulls every subsequent estimate toward it - even when the anchor is irrelevant or random.
Open conceptBase rate (and base-rate neglect)
The background frequency of something across the relevant population - and the chronic mistake of ignoring it in favour of vivid specifics.
Open conceptPlanning fallacy
We chronically underestimate how long things will take, how much they will cost, and how much can go wrong.
Open conceptOutside view
Estimate by treating your case as a member of a reference class of similar past cases - not by reasoning from inside your specific story.
Open conceptInside view
Estimating a case by reasoning from inside its own story and details - usually too optimistic, usually overconfident.
Open conceptLoss aversion
Losses hurt about twice as much as equivalent gains feel good. We will work harder to avoid losing $100 than to gain $100.
Open conceptFraming effect
How a choice is described changes which option people pick - even when the underlying outcomes are mathematically identical.
Open conceptRegression to the mean
Extreme observations are usually followed by less extreme ones, regardless of any intervention. Mistake regression for causation and you'll see patterns that aren't there.
Open conceptOverconfidence
We are systematically more certain than we should be, especially about predictions and our own judgment.
Open conceptWhen expert intuition can be trusted
Two conditions: a regular, predictable environment AND fast, clear feedback. Only when both hold can intuition become genuinely reliable.
Open conceptNarrative fallacy
We turn messy reality into clean stories. Stories explain in hindsight but predict almost nothing.
Open conceptBlack Swan
A rare, high-impact event that is unpredictable beforehand and feels obvious afterwards.
Open conceptMediocristan
Domains where no single observation dramatically changes the total - physical attributes, repeating natural processes, well-bounded quantities.
Open conceptExtremistan
Domains where a single observation can dominate the total - wealth, book sales, pandemic spread, social-media reach, war casualties.
Open conceptThick tail (and thin tail)
In thick-tailed distributions, extreme events are more frequent and more extreme than the bell curve predicts. The tail does most of the work.
Open conceptSilent evidence
The data you don't see - failures, drop-outs, missing cases - usually changes the conclusion. The story you can hear is biased by survivorship.
Open conceptSurvivorship bias
Drawing conclusions only from the survivors of a selection process, ignoring everyone the process eliminated.
Open conceptRuin
The class of outcomes from which you cannot recover. Ruin doesn't average out, doesn't get reversed by a winning streak, and ends the game.
Open conceptExposure
The shape of what happens to you under different outcomes - not the probability of those outcomes. Manage exposure, not forecast.
Open conceptOptionality
The asymmetric right (not obligation) to take advantage of good outcomes without being trapped by bad ones.
Open conceptBarbell strategy
Combine extremely safe positions with extremely risky ones. Avoid the «moderate» middle, which has thin-tail-thinking risk you can't see.
Open conceptPositive asymmetry (convexity)
Shape your bets so the downside is small and the upside is large. The asymmetry - not the probability - is what generates positive long-run results.
Open conceptSkin in the game
Trust advice and decisions from people who pay the cost of being wrong. Distrust those who don't.
Open conceptDoxastic rules
Belief-handling rules: track confidence as a number, update it on evidence, and gate big actions behind high p.
Open conceptEpistemic arrogance
Confidence that vastly exceeds the quality of evidence - usually because the speaker can't see what they don't know.
Open conceptAntifragility
Not the opposite of fragile. Beyond robust. Antifragile systems benefit from stress, variation, error, and time.
Open conceptFragility / Robustness / Antifragility triad
Three shapes under stress: fragile breaks, robust holds, antifragile improves. Most systems aren't classified - they should be.
Open conceptVia negativa
Improve by subtracting harm - not by adding clever fixes. Removing bad inputs is more reliable than adding good ones.
Open conceptSingle point of failure
A component whose failure brings down the whole system. The classic structural fragility. Almost always cheaper to remove than to defend.
Open conceptSmall local failures (design for them)
Build systems where small components fail often, cheaply, and visibly - so the whole system gets information without ever risking ruin.
Open conceptLindy effect
For non-perishable things (ideas, books, technologies), the longer something has been around, the longer it's likely to stick around.
Open conceptNeomania
Falling for newness as if it were a virtue. The neomaniac upgrades for the sake of upgrading.
Open conceptIatrogenics
Harm caused by the intervention itself. Treatment that's worse than the disease.
Open conceptIntervention bias
The compulsion to «do something» even when doing nothing has the better expected outcome.
Open conceptFragilista
Someone whose actions reliably make systems more fragile while sounding smart. The well-meaning destroyer.
Open conceptConvexity
A position is convex when small increases in input produce disproportionate gains, while losses are bounded.
Open conceptEpilogism
Reasoning by pattern from concrete observable cases, refusing to extrapolate beyond what's directly seen.
Open conceptAmor fati
«Love of fate». Embrace what happens - including loss, pain, and reversal - as material for the system you're building.
Open conceptPractical templates
Tail Risk Scan
Map where you're exposed to outcomes you can't recover from.
Via Negativa Checklist
Find the cheapest improvement: what to remove instead of what to add.
Bias Scanner
Before a decision, surface the most likely cognitive shortcuts that are running the show.
Outside View Planner
Re-estimate a plan by treating it as a member of a reference class instead of a unique case.
Barbell Strategy Planner
Restructure a position (career, portfolio, project) into an extreme-safe + extreme-upside combo.
Single Point of Failure Audit
Find the components whose failure would bring everything down.
Doxastic Confidence Update
Track and update p - your belief level in a hypothesis - as evidence arrives.
Antifragility Quick Check
30-second triage: is this system fragile, robust, or antifragile right now?
Concept comparisons
Fragile vs Robust vs Antifragile
Any system you can't easily rebuild. Career, relationships, supply chain, codebase.
Open comparisonMediocristan vs Extremistan
Whenever you plan to «average» something. First classify the domain.
Open comparisonInside View vs Outside View
Every project estimate, every revenue forecast, every commitment.
Open comparisonAvailability vs Base Rate
Risk estimation, hiring, forecasting, anything where vivid recent memories influence judgment.
Open comparisonWYSIATI vs Silent Evidence
Any decision based on success stories, expert testimony, or curated examples.
Open comparisonOptionality vs Forecasting
Any high-uncertainty bet: career, product, market timing. Default to optionality.
Open comparisonThis part of the knowledge base is inspired by the book. Go to the Ukrainian edition to explore the concept in depth.