From Polls to Markets
Election prediction markets had a notable run during the most recent US presidential cycle, at times appearing to lead traditional polling aggregators. Damodaran's analysis is a useful opportunity to sit with a broader methodological question: when should we trust crowd-sourced market prices as forecasts, and when should we distrust them?
Polls are individually noisy. Aggregators reduce that noise, but they still rely on opt-in respondents answering a survey question. Prediction markets substitute a different signal: real-money bets by participants who face a direct financial consequence for being wrong. The two mechanisms have different failure modes.
The Wisdom of Crowds
The case for crowd-based forecasts is well established. Consumers routinely trust aggregated judgments — movie ratings, product reviews, star scores — and these collective assessments often outperform expert opinion.
Prediction markets extend the same logic to probability estimation:
- Market prices aggregate dispersed information held by many participants
- Prices update continuously as new information arrives
- Participants stake real capital, which disciplines overconfidence
When those conditions hold, market-implied probabilities can be powerful forecasts.
The Madness of Crowds
Crowds are not always wise. Damodaran points back to the 1841 classic Extraordinary Popular Delusions and the Madness of Markets for a reason — financial history is littered with cases where aggregated market judgment went badly wrong.
Failure modes to watch for:
- Noise trading that swamps informed participation
- Momentum effects that drive prices away from fundamentals
- Participant bias when the betting population is not representative
- Market manipulation by large or coordinated players
- Illiquidity that lets small flows move prices disproportionately
The question of whether a given prediction market is wise or mad at a given moment depends on which forces dominate.
Using Both Market and Expert Signals
Damodaran's recommendation is to avoid the either/or framing. Market prices and expert forecasts measure different things and have different biases. The useful practice is:
- Use market prices as a real-time, information-aggregating baseline
- Use expert analysis to interrogate whether the market signal is corrupted by noise, manipulation, or unrepresentative participation
- Weight them based on the specific event being forecast
Valuation Implications
The same logic applies inside valuation work. When a company has a traded stock, the market price is a real-time crowd forecast of its value. Ignoring that price is naive, but treating it as automatically correct is equally naive. The practitioner's job is to identify which signal is more reliable for the specific question being asked — and to be honest about when the market is more informed than the analyst, which is often.
Bottom Line
Prediction markets are a useful forecasting tool, not an oracle. The wisdom of crowds is real and powerful when participation is broad and informed; the madness of crowds is equally real when it isn't. The discipline is knowing which regime you are in.