Correlated Positions That Look Diversified
Twenty positions on six underlying events is six bets. Spotting hidden correlation in an event market portfolio, where it hides in plain sight.
A portfolio of twenty event contracts feels diversified. If they resolve on six underlying facts, it is six positions, and its risk is several times what the count suggests.
Where the correlation hides
Strike ladders. Several thresholds on the same figure are points on one distribution. If the figure comes in high, every one of them resolves accordingly. This is the most common case and the least noticed. How a contract is put together.
Multi-outcome sets. Holding several candidates in one field is one position with a known combined payoff, not several bets. Multi-outcome markets.
Shared drivers. Inflation contracts, rate decision contracts and related market-level contracts all move on the same releases. They are different markets and largely one exposure. Interest rate markets.
Same-day events. Several contracts resolving from one election, one meeting, one match.
Shared venue. Every position on one platform shares that platform's operational and regulatory exposure. Counterparty and platform risk.
Shared resolution mechanism. Positions settled by the same oracle process share whatever risk that process carries. Disputes.
Shared reasoning. The subtlest. If every position came from the same model or the same source, they share one error. This does not appear in any correlation matrix computed from prices.
The specific cross-venue case
A two-leg pair across venues is designed to be perfectly negatively correlated — one wins, one loses, and the difference is the edge.
That design fails in exactly the situations that matter: the contracts resolve differently, or one voids while the other settles. Then you hold two positions that both lose, and the loss is the whole position rather than the few cents the pair was meant to earn.
So the correlation of a hedged pair is not minus one. It is minus one most of the time and something much worse occasionally, and the occasional case dominates the expected value. Resolution differences.
Measuring it
Forget correlation coefficients computed from price series; the sample is too short and the structure matters more than the statistics.
Instead, group positions by underlying fact. Every position that resolves from the same observation goes in one group. Then ask, for each group, what happens if it goes the wrong way. That is your real exposure map, and it takes ten minutes.
Managing it
Limit exposure per underlying event, not per contract.
Count committed capital by group. Ten positions on one election is one position of that total size.
Treat a strike ladder as a single position with a shaped payoff.
Diversify the reasoning, not only the tickets. Positions from one model are one bet on that model.
Include the venue as a grouping dimension. It is a real shared exposure even when the questions are unrelated.
Why it matters most here
Because event markets encourage it. Listing a dozen thresholds on the same figure makes it natural to hold several. Each feels like a separate decision, and the platform presents them as separate rows.
They are not separate. Counting them as separate is how a portfolio that looks conservative turns out to have been one large bet all along. Position sizing and variance and drawdown.
Next
Educational content only. Nothing here is financial advice, a solicitation, or an offer to trade. Prediction markets can move against you and a position can lose its entire value. Where this guide describes a difference between venues it describes a theoretical cross-market edge in observed prices, which is not a profit: fees, spreads, capital lockup, transfer time and differing settlement rules all come out of it first.
Keep reading
Risk and Costs · 3 min readVariance, Drawdown, and What a Losing Run Feels LikeWith a small genuine edge, long losing stretches are normal and indistinguishable from having no edge at all. Planning for that in advance is the whole game.
Risk and Costs · 3 min readWhen Not to TradeThe default action in a market with real costs is no action. A checklist of the conditions under which the correct response to an interesting number is nothing.
Risk and Costs · 3 min readThe Behavioural Traps Specific to Event MarketsBinary payouts, visible probabilities and questions you have opinions about combine into a specific set of psychological failure modes.
What Markets Price · 3 min readInflation Print Markets and the Detail That Decides ThemWhich index, which vintage, which decimal. Inflation contracts are settled by details that most participants never read, and occasionally that matters.