Alternatives & Private Markets

Prediction Market

Also known as: Event contract, Event derivative

A contract paying $1 if an event happens and nothing otherwise, so its price reads as a probability. A forecasting instrument that is also, unavoidably, a wagering one.

4 min read · 778 words

Asset class
Alternatives (event-linked)
Instrument type
Binary event contract
Traded
Regulated exchanges in some jurisdictions; offshore elsewhere
Typical users
Forecasters, hedgers of event risk, speculators
Stylised payoff at expiry (not to scale).
KDigital callSpot at expiryPayoff
1 · SnapshotThe one idea to remember
Key intuition: the price is the probability. Everything worth arguing about is whether that probability is well-calibrated — and it is systematically less reliable at the extremes.
2 · BeginnerWhat is it, really?

A prediction market lists a contract that settles at $1 if a specified event occurs and $0 if it does not. Because the payoff is fixed, the price carries a direct interpretation: a contract trading at 0.63 means the market prices the event at roughly 63%.

This is the same instrument as a binary option, with two differences that matter:

  • The price is set by an order book, not quoted by a counterparty taking the other side of your trade.
  • The underlying is an event — an election, an economic release, a regulatory decision — rather than a market price.

The interesting claim is not that individuals forecast well. It is that a market aggregates dispersed information and forces participants to back opinions with money, which filters out talk. Whether it does that better than alternatives is an empirical question with a genuinely mixed answer.

3 · IntermediateHow it works in practice

Where the price stops being a clean probability

  • The longshot bias. Contracts at very low prices trade persistently above their realised frequency. A 3% contract has historically resolved yes less often than 3% of the time — the same bias documented for decades in betting markets, and it survives in prediction markets too.
  • The cost of capital. Buying a contract at 0.95 to make 0.05 ties up capital until resolution. Over six months that is an unattractive annualised return, so nobody arbitrages the last few cents — which is exactly why extreme prices are least reliable.
  • Fees and spreads sit inside the price. A market quoted 0.62/0.65 has no single probability, and the mid is a convenience.
  • Thin markets are opinion, not aggregation. A contract with little volume reflects whoever showed up, and the informational claim rests entirely on depth.

Genuine hedging uses

ExposureEvent contract as a hedge
A regulated firm facing a rule changeBuy the contract on the rule passing
A farmer facing weatherOverlaps with weather derivatives, at retail size
A business exposed to a policy outcomeCheaper than restructuring operations in advance

These are real, and they are also the argument regulators have found least persuasive — because the same contract serves a hedger and a gambler identically, and the venue cannot tell them apart.

Worked example: a contract on an economic threshold trades at 0.40 with a 0.02 spread. Buying at 0.41 to receive 1.00 is a 144% gross return if correct — and if the true probability is 40%, the expected value is 0.40 × 1.00 − 0.41 = −0.01. The spread alone turns a fairly priced market into a negative-expectation trade.
4 · AdvancedPricing & valuation

The evidence on accuracy, honestly summarised

  • Prediction markets have generally matched or modestly beaten polling averages for election outcomes, and have done well on binary, well-defined, near-dated questions with liquid markets.
  • They have done poorly on long-dated questions, on questions with ambiguous resolution criteria, and at extreme probabilities.
  • They are not magic aggregators: they reflect the beliefs of the people trading them, weighted by capital. A market dominated by one demographic prices that demographic's beliefs, and the money-weighting is a feature only if the informed are also the well-capitalised.
  • The most robust finding is comparative rather than absolute: a market price is a hard-to-beat baseline, and beating it consistently requires genuine private information — the same standard as any other market.

The regulatory position is the product's main risk

Jurisdictions differ sharply, and the differences are not stable:

  • Some regulators have authorised event contracts on designated exchanges with position limits and defined categories.
  • Others treat them as gambling, subject to entirely separate licensing, or prohibit them outright.
  • Contracts on elections have been the most contested category, litigated in several jurisdictions with outcomes that have shifted more than once.
  • Offshore and crypto-settled venues operate outside all of this, and carry counterparty and settlement risks that no regulated exchange would permit — the custody question, again.

Resolution risk is the underrated one

Every contract depends on an agreed source declaring the outcome. Ambiguous wording, a delayed source, a revised statistic or a disputed result creates genuine settlement risk on a contract whose whole appeal is that the payoff is unambiguous. Serious venues publish detailed resolution criteria for exactly this reason, and reading them is the diligence — the payoff structure has no subtlety, and the definition has all of it.

The formulas above are standard textbook formulations, simplified for teaching. They explain the mechanism — they are not a valuation tool, and they will not reproduce a dealer’s price.

5 · Desk notesHow practitioners think about it
Practitioner note: read the resolution criteria before the price. Then check the depth: a probability derived from a market with no size behind it is one person's opinion with a decimal point attached.