Bayesian Thinking is a framework for updating beliefs in proportion to evidence. Named for the Reverend Thomas Bayes, whose theorem formalizes the mathematics of belief revision, it provides a principled method for incorporating new information into prior beliefs — neither overreacting to single data points nor clinging to existing views against contradicting evidence. It is the foundation of modern statistics, machine learning, and rational decision-making under uncertainty.
The Inside View vs. Outside View is a conceptual distinction developed by Daniel Kahneman and Amos Tversky that describes two modes of forecasting. The Inside View uses the specific details of a situation — plans, capabilities, intentions — to build predictions. The Outside View consults the base rate of outcomes for comparable past situations. The Inside View produces more confident, more optimistic forecasts; the Outside View produces more accurate ones. The distinction explains why most plans fail to anticipate obstacles and why reference class data is more reliable than expert case analysis.
Reference Class Forecasting is a method of estimation and prediction developed by Nobel laureate Daniel Kahneman and Amos Tversky that deliberately anchors predictions to observed base rates from comparable past projects or situations, rather than relying on case-specific analysis. By forcing forecasters to consult the 'outside view' — the statistical distribution of outcomes for similar situations — it corrects for the systematic optimism bias and inside-view thinking that causes most projects to run over time and over budget.