Part II: Market Microstructure & Pricing
Chapters 7--12
If Part I taught you what prediction markets are, Part II teaches you how they actually work beneath the surface. The prices you see on a prediction market platform are not handed down by some oracle; they emerge from a complex interplay of order flow, mechanism design, and information dynamics. Understanding this machinery is what separates a casual observer from someone who can reason precisely about why a price moved, whether it moved enough, and what structural forces might push it further.
Market microstructure is a field that originated in traditional finance, studying how exchanges translate the intentions of buyers and sellers into executed trades and observable prices. Prediction markets inherit many of these ideas but also introduce novel wrinkles -- thin markets, bounded contracts, and automated mechanisms that replace the human market makers of Wall Street with elegant mathematical functions. This part covers the essential microstructure concepts, adapted and extended for the prediction market setting.
Chapter 7 begins with the order book: the fundamental data structure that organizes the bids and asks in a market. You will learn how limit orders, market orders, and order matching algorithms determine which trades execute and at what price. Even on platforms that do not expose a traditional order book, understanding this model provides the right mental framework for reasoning about liquidity and price impact.
Chapter 8 introduces automated market makers, arguably the most important innovation in prediction market design. We study the Logarithmic Market Scoring Rule (LMSR) in depth -- its cost function, its properties, and why Robin Hanson's invention proved so influential. We then examine the Constant Product Market Maker (CPMM) popularized by decentralized finance, comparing its strengths and weaknesses against the LMSR. By the end of this chapter, you will be able to calculate the cost of a trade, the resulting price shift, and the liquidity parameter's effect on market responsiveness.
Chapter 9 broadens the lens to scoring rules in general. Proper scoring rules are the mathematical backbone that ensures participants are incentivized to report their true beliefs. We cover the Brier score, the logarithmic score, and the deep connection between scoring rules and market maker mechanisms. This chapter is more theoretical than most, but its payoff is a crisp understanding of why prediction markets work as well as they do -- and under what conditions they might fail.
Chapter 10 addresses the frictions that every trader encounters: transaction costs. Fees, spreads, slippage, and price impact all eat into expected returns. We quantify these costs, show how they vary across platforms, and demonstrate how to incorporate them into any trading or modeling decision. Ignoring transaction costs is the fastest route to a strategy that looks profitable on paper but bleeds money in practice.
In Chapter 11, we tackle the central theoretical promise of prediction markets: information aggregation. Under what conditions do market prices efficiently aggregate the private information of diverse participants? We examine the efficient market hypothesis as it applies to prediction markets, review the empirical evidence on forecast accuracy, and discuss the well-known limits -- thin markets, manipulation, and correlated information. You will come away with a realistic, evidence-based view of when you should trust a market price and when you should be skeptical.
Chapter 12 ties these threads together with the concept of calibration. A well-functioning prediction market should produce prices that, over many events, correspond closely to observed frequencies: events priced at 70% should happen roughly 70% of the time. We show how to measure calibration, diagnose miscalibration, and understand its sources. Calibration analysis is not just an academic exercise; it is one of the most powerful tools for identifying exploitable inefficiencies.
By the end of Part II, you will understand the mechanics of price formation, the incentive structures that make markets work, and the frictions that create both challenges and opportunities. This knowledge is the analytical engine that powers everything in Parts III and IV -- you cannot trade effectively or build models wisely without it.