Event-Driven Trading on Polymarket: Positioning Before the Catalyst

Event-driven trading means taking a position because a scheduled event will force a market to reprice: an election, a court ruling, a central-bank decision, a data release, an earnings date. On Polymarket the trade carries three separate risks — the outcome, the repricing, and the resolution — and most traders price only the first.

That omission is expensive. You can be right about the world, right about the direction, and still lose money because the market did not converge fast enough, because the exit was illiquid, or because the resolution rule meant something different from the headline.

The calendar is the strategy

Start with dates, not opinions. Every market has an end date and a resolution source, and the rules define how ambiguous cases are handled. Polymarket's own documentation makes the point directly: the market title describes the question, but the rules define how it resolves, and traders should read them before trading.

A single news event often produces several markets with different cutoffs, different sources, and different wording. "Will the bill pass by Friday" and "Will the bill become law this year" can trade at very different prices even though they sound similar. That spread is an opportunity only if you have read both rules; otherwise it is a trap.

Build the calendar in advance. For each event, record the scheduled time, the resolution source, the exact criterion, and the sibling markets that resolve on the same facts. The trade usually comes from the gap between the rule and the headline, not from a general view about the event.

Two different pre-event trades

There are two ways to position before a catalyst, and they have different economics.

Hold to resolution. You buy at price c, believe the true probability is p, and intend to hold until the contract pays $1 or $0. Your edge is p minus c, minus fees. You are paid for being right about the world.

Exit on repricing. You buy before the event and plan to sell after the market updates, without waiting for resolution. Your edge depends on where the price settles and how quickly, not just on whether the event happens.

The second trade is more popular and less forgiving. A market can move most of the way to your view and still leave you with nothing after fees.

A worked pre-event position

Suppose a contract trades at 0.62 before a scheduled ruling. You estimate the true probability at 0.70 and buy 1,000 shares at the 0.64 ask.

The entry fee is 1,000 × 0.04 × 0.64 × 0.36, or $9.22, using the politics taker rate.

If you hold to resolution and the outcome happens, you receive $1,000. Profit is $1,000 minus $640 minus $9.22, or $350.78. If it does not happen, you lose $649.22. At your 70% estimate, expected value is 0.70 × 350.78 minus 0.30 × 649.22, or about $50.78, roughly 7.8% of the $649 stake.

If instead you plan to exit when the market reprices to 0.70, the gross gain is $60. Exit fees at 0.70 are 1,000 × 0.04 × 0.70 × 0.30, or $8.40, so net profit is $42.38, about 6.5%. If the market only reprices to 0.66, the gross gain is $20 and round-trip fees are about $18.20, leaving under $2. You were right about the outcome and made almost nothing.

That is the repricing trade in one line: convergence has to overshoot your transaction costs, not merely reach fair value.

Resolution risk is a separate bet

Polymarket resolves prediction markets through the UMA optimistic oracle. Anyone can propose an outcome by posting a bond, typically $750. There is a two-hour challenge period. If a proposal is disputed, a second round begins; if that is disputed too, the question goes to a UMA token-holder vote, and the full process can take four to six days.

The documented outcomes include a 50/50 resolution when the event is genuinely ambiguous. In that case each token redeems for $0.50. A position bought at 0.90 loses 40 cents per share even though no one was "wrong" about the news.

Two practical consequences follow.

First, resolution risk is not uniform. A market resolved by an official government release is different from one resolved by "credible reporting" or a single named website. The more discretion in the rule, the wider the range of outcomes the market can produce, and the more the price should compensate you for that.

Second, the resolution process has its own clock. Winning positions do not become cash the moment the event ends. Trading stops at resolution, winning tokens redeem for $1, and losing tokens for $0. If you need liquidity before that, you are selling to someone else at a discount, not redeeming.

Hedging the event

Polymarket's negative-risk mechanism is the cleanest hedging tool for multi-outcome events. In an event where only one outcome can win, one No share in any market can be converted into one Yes share in every other market. Betting against one outcome is economically the same as betting on all the others, and the conversion lets you express that without buying each leg separately.

For cross-market hedges, the binding constraint is rule consistency. Two contracts that look complementary may have different resolution sources or different deadlines. If they do, you have a hedge in name only. You have also paid two spreads for the privilege.

Hedging is not free even when the rules match. You cross the spread on both legs, and the hedge ratio is not one-to-one unless the payoffs are exactly complementary. In practice, a hedge reduces variance and caps loss; it does not eliminate the position.

Post-event liquidity

After the event, trading usually continues until resolution, but the character of the book changes. Market makers who were quoting before the catalyst often widen or leave; the remaining liquidity is directional. If you planned to exit on repricing, this is the moment you need it most.

Check the book before the event, not after. If the exit depends on a bid that does not exist yet, your plan is a hope. A useful test: assume you can only sell at the current bid, minus one tick, and recompute the trade. If it is still attractive, the plan is robust.

Sizing and a rule set

Event positions should be sized for a total loss, because a binary outcome does not stop out. A gap from 0.60 to 0.05 is a single print, and no stop order can execute inside it.

1. Read the resolution rule and identify the exact source before sizing.
2. Write the falsifier: the specific fact that would prove you wrong.
3. Decide in advance whether you are holding to resolution or exiting on repricing. Do not switch after the fact.
4. Cap exposure per event, not just per market. Three markets on the same election are one bet.
5. Check sibling markets for a hedge before the event, and verify that their rules match.
6. Record resolution surprises. If a market resolved differently from your reading, that is a process failure worth reviewing even if the trade made money.

What the edge actually is

The edge in event-driven trading is rarely a superior forecast of the event itself. Large, public events are heavily covered, and the price usually reflects the consensus. The edge is more often in the details: a rule that is narrower than the headline, a market that has not attracted attention yet, a resolution source that is more definitive than the crowd assumes, or a hedge that is cheaper than the risk it removes.

Those edges are small, and they are eaten by fees and spread. Which is why the discipline is mostly about saying no: no trade without a written rule, no position without a pre-committed exit, and no size that turns a forecast into a solvency question.