Part I: Foundations
Chapters 1--6
Every discipline has its bedrock -- the core ideas that, once understood, make everything else click into place. For prediction markets, that bedrock is surprisingly rich. It draws from economics, probability theory, computer science, and behavioral psychology, and it weaves these threads into something genuinely new: a mechanism that turns the dispersed knowledge of many individuals into a single, actionable number.
This first part of the book is devoted entirely to building that foundation. If you are tempted to skip ahead to trading strategies or machine learning models, resist the urge. The concepts introduced here will recur on every subsequent page, and a shallow understanding of them is the single most reliable predictor of costly mistakes later on.
We begin in Chapter 1 with the most fundamental question: what exactly is a prediction market? You will learn how these markets differ from traditional financial exchanges, from opinion polls, and from expert panels. We will pin down a precise definition and explore the key claim that makes prediction markets fascinating -- that market prices, under the right conditions, can serve as well-calibrated probability estimates of future events. This chapter also introduces the vocabulary you will use throughout the book: contracts, shares, resolution, and the notion of a market as an information aggregation device.
Chapter 2 traces the surprisingly deep history of prediction markets. From papal election betting in sixteenth-century Italy to the Iowa Electronic Markets and the ill-fated DARPA Policy Analysis Market, the story of prediction markets is one of recurring invention, political controversy, and hard-won empirical validation. Understanding this history is not mere trivia. It reveals the design choices and regulatory constraints that shape every platform operating today.
In Chapter 3, we turn to the mathematical language that prediction markets speak: probability theory. You do not need a graduate degree in statistics to follow this material, but you do need fluency in the basics -- sample spaces, conditional probability, Bayes' theorem, and the concept of calibration. This chapter provides that fluency, with every concept grounded in prediction market examples rather than abstract coin flips.
Chapter 4 dissects the mechanics of prediction market contracts themselves. What does it mean to buy a "Yes" share at 63 cents? How are contracts created, traded, and resolved? What happens when the resolution criteria are ambiguous? We will examine binary contracts, multi-outcome contracts, and scalar (range) contracts, and you will develop an intuition for how contract design influences the information a market can capture.
Chapter 5 surveys the current platform landscape. From Polymarket and Manifold Markets to Metaculus and the legacy platforms like PredictIt, each venue makes different design trade-offs around fees, liquidity, regulation, and user experience. By the end of this chapter, you will know where to look for the markets most relevant to your interests and how to evaluate a platform critically.
Finally, Chapter 6 gets your hands dirty. We set up a Python development environment tailored to prediction market analysis -- installing key libraries, connecting to public APIs, and pulling your first batch of live market data. This chapter is deliberately practical: by its final page, you will have a working notebook that fetches real prices and plots a basic probability time series.
Together, these six chapters give you the conceptual vocabulary, historical context, mathematical toolkit, institutional knowledge, and technical environment you need to engage with prediction markets at a serious level. Think of Part I as the map and the compass. The territory -- trading, modeling, market design, and beyond -- awaits in the parts that follow.