Polymarket Arbitrage: Prove the Payout Identity

A Polymarket Arbitrage Payout Identity Has Conditions

An identity is stronger than a forecast. A forecast says an outcome is likely; an identity says how the contract terms and positions combine regardless of which covered outcome occurs. If a claim depends on assuming that an unlikely event will not happen, it is not an unconditional coverage statement. That distinction matters even when the neglected event has never occurred in the analyst's sample.

Write the identity in ordinary language before using a scanner. For a standard complete binary set, equal quantities of Yes and No refer to the same condition and the permitted combined payout is the collateral amount represented by the set. The same words applied to two different conditions may be false. Identity requires matching the actual condition, quantities and applicable settlement or conversion mechanics.

Polymarket's position documentation describes split and merge operations for complete sets, with current International documentation using pUSD collateral. Those mechanics are specific to the supported tokens and account context. They should not be assumed for arbitrary claims or for Polymarket US, which has its own product and account environment.

The older Polymarket arbitrage bot article provides background on complete-set economics. A bot can help maintain the records, but it cannot make a false identity true. The analyst still needs to identify the exact assets and explain why the proposed operation returns the amount used in the calculation.

Work Through a Polymarket Arbitrage Complete Set

Assume a fictional standard market has executable asks of $0.44 for Yes and $0.53 for No, with 100 units available on each side. Acquiring 100 of each costs $44 plus $53, or $97 before charges. If the tokens form a verified complete set redeemable or mergeable for $100 under the relevant mechanics, the gross difference is $3.

Now assume $1.40 of total acquisition and completion costs for the illustration. The all-in amount is $98.40 and the difference becomes $1.60. That is approximately 1.63% of the $98.40 committed. The calculation is not a quoted fee schedule, and it does not show that these prices or costs are available now. It demonstrates which quantities need to be compared.

The gross difference and the completed account result are separate records. Before both legs execute, the account does not own the planned complete set. Before a merge or final payout is confirmed, the expected collateral return has not yet become a verified cash flow. Calling the initial displayed gap realized profit skips the steps that establish the result.

Also check quantity units. Equal dollar spending does not generally buy equal outcome quantities when prices differ. Spending $44 on each side at the assumed prices buys 100 Yes and about 83.02 No, before precision constraints. Only matched quantities form complete pairs; the remaining Yes units retain directional exposure. A coverage calculation should pair units, not balances spent.

Polymarket Arbitrage Related Outcomes Need a Test

Consider two fictional contracts about an indicator exceeding 60. One uses the first official publication; the other uses a revised value released a week later. Buying Yes on the first and No on the second may look complementary on a headline screen, but the two claims can both lose. The first value could be below 60 while the revision rises above it.

They can also both win if the first value exceeds 60 and the revision falls below it. Those possibilities prove that the combined payout is variable. A small combined acquisition cost might still interest a directional researcher, but the position requires a forecast about revisions. It should not inherit the complete-set label merely because the displayed titles contain the same threshold.

Time definitions create another mismatch. A claim about a price at noon differs from a claim about whether that price is reached at any time before noon. A path that rises above the threshold at 11:00 and falls below it by noon produces different outcomes. The comparison must preserve the observation rule as well as the asset name and threshold number.

The related Kalshi arbitrage strategy guide extends this discipline into another venue context. Cross-venue similarity is a hypothesis to verify, not evidence that two claims share a payout. Matching the words on a screen is only the beginning of the contract review.

Mutually Exclusive Does Not Mean Exhaustive

Outcomes are mutually exclusive when no two can occur together under the relevant rules. They are exhaustive when at least one must occur. A complete collection of ordinary one-dollar outcomes needs both properties to return exactly one dollar per equal-quantity set. Satisfying just one leaves either overlap or a coverage gap.

Imagine three fictional winner contracts for North, South and East. If exactly one of those three must be the official winner, equal Yes quantities cover the ordinary winner cases. If West can win but has been omitted, the three contracts remain mutually exclusive yet are not exhaustive. Buying all three can produce a zero payout when West wins.

Threshold markets illustrate the opposite issue. Claims that an index exceeds 50, 55 and 60 are not mutually exclusive. A value of 62 makes all three true. Their prices should not be added as though they describe disjoint ranges. Turning thresholds into a range structure requires careful reasoning about combinations and boundaries, not merely collecting every market under one event heading.

A useful review writes out a concrete counterexample whenever possible. If the analyst cannot explain what happens at a boundary, during a cancellation or when an omitted outcome occurs, the coverage statement remains incomplete. A convincing-looking sum cannot substitute for this case analysis.

Read Special Conversion Structures as Special Structures

Some supported event designs provide conversions among outcomes that do not exist for unrelated markets. Polymarket's negative-risk documentation describes such relationships for designated multi-outcome events. The key practical distinction is that the relationship depends on the event's implemented structure, not simply on the analyst's belief that one outcome should win.

A scanner should verify the supported structure and exact identifiers before treating a conversion as available. An economic equivalence on paper and an executable conversion in an account are different facts. The account may need the correct token type, permission and transaction path, and the actual operation must be reconciled after submission.

Changing outcome definitions require additional care. An Other category or a placeholder can have meaning that evolves under the event's documented process. A historical mapping should retain the definition reviewed at the time. If a candidate relies on a set of named outcomes that no longer describes the supported structure, refresh the analysis before making a new commitment.

This is also why a generic rule such as buy all prices totaling below one can be unsafe. The list shown in an interface may be incomplete, filtered or include outcomes with special treatment. The analyst needs the actual outcome universe and its rules, with a clear explanation of why the intended collection has the claimed return.

Price the Quantity That Can Actually Be Acquired

Return to the fictional binary market, but now assume only 40 No units are available at $0.53 and the next 60 cost $0.56. Acquiring 100 No costs $21.20 plus $33.60, or $54.80. Add the $44 Yes leg and the pre-charge cost becomes $98.80. The original $3 gross gap has narrowed to $1.20 before any charges.

If total costs remain the hypothetical $1.40, the completed set costs $100.20 against $100 of proceeds. The candidate has a $0.20 loss under those assumptions. This is not a failure of the payout identity; it is a failure of the acquisition economics. Keeping those diagnoses separate helps the analyst improve the right part of the process.

Depth should be observed at compatible times. Combining yesterday's cheap Yes ask with today's cheap No ask does not establish that both could be acquired together. Record the time and source of each book. Even simultaneous observations are not reservations, so a candidate needs price bounds and an execution policy in addition to an estimate.

The older how to arbitrage Polymarket guide develops the sequence of acquisition and recovery. Its relevance is practical: the cost calculation needs to survive the path from displayed book to actual fills. A spreadsheet total is only a candidate until the account owns the required quantities at verified prices.

Partial Execution Changes the Nature of the Position

Suppose 100 Yes units fill at $0.44, but only 40 No units fill at $0.53. Before charges, the account spends $65.20. Forty matched pairs account for $38.80 of that cost and correspond to $40 under the complete-set assumption. The remaining 60 Yes units cost $26.40 and have an outcome-dependent payout.

If Yes wins, total ordinary payout is $100: the 100 Yes units pay and the No units do not. If No wins, total payout is $40. Against the $65.20 acquisition cost, those branches produce $34.80 or a $25.20 loss before charges. The account has not earned the planned complete-set return merely because one leg filled successfully.

A recovery decision should compare current completion cost, current exit proceeds and the remaining permitted exposure. It should not automatically pursue the second leg at any price in order to restore the original label. Completing a hedge can crystallize a known loss; leaving it incomplete preserves directional risk. The permitted choice belongs in the strategy before the incident occurs.

The related Polymarket arbitrage strategy article turns this analysis into execution bounds. An identity is a statement about a completed position. A strategy also needs instructions for the states that arise before completion, including partial fills, rejected legs and unknown submission results.

Acknowledgment Does Not Establish the Required Holdings

An order response can establish that a venue accepted an instruction without establishing that every intended unit filled. A delayed or resting order remains different from a completed trade. Preserve order identity, filled quantity and remaining commitment so the account can distinguish actual holdings from possible future holdings.

Order type changes the meaning of a remainder. An immediate partial-fill order with cancellation of the unfilled balance does not leave that balance resting. Conversely, a persistent order may continue to create exposure after the first partial execution. Interpret the actual order type instead of using one generic partial status for every case.

If a submission result is unknown, investigate the original attempt before creating a fresh independent order. The first attempt may already have produced a fill. Reserving the maximum possible commitment during uncertainty helps prevent the account from allocating the same capital to another candidate while the original exposure remains unresolved.

The older prediction market arbitrage software discussion explains why software evaluation must include these intermediate states. A tool that reports theoretical spread well but loses track of actual legs can turn a structurally valid idea into an uncontrolled account position.

Completion Time Belongs in the Economic Comparison

A fixed eventual payout does not mean the capital is immediately available. A supported merge may provide one completion path; waiting for resolution may provide another. Each has its own actual costs, operational dependencies and timing. Compare the feasible path for the specific account rather than assuming that every candidate releases funds at the same speed.

Suppose the fictional $1.60 net difference requires $98.40 for two days in one case and for twenty days in another. The nominal result is identical, but the capital commitment differs. A simple annualized extrapolation would be misleading if it assumes the same opportunity can be repeated continuously without gaps, costs or capacity limits.

Capital also has a location. Funds available in one account or form may not be usable for the other required action. A transfer estimate should not become a confirmed balance until the account can actually use it. Include unresolved transfers and reservations when determining whether both legs can be completed within the intended limits.

This treatment of time is not an argument that every slow candidate is unsuitable. It gives the reader a consistent comparison: amount committed, possible recovery paths, expected holding period and uncertainty about release. A small nominal difference can be unattractive when it consumes scarce operating capacity for an unpredictable interval.

Separate Payout Risk From Operational Risk

A verified complete-set identity can remove ordinary event-direction uncertainty for the matched quantity while leaving other dependencies. The account still relies on correct asset identification, supported conversion or settlement, functioning infrastructure and accurate custody records. Describing the economic identity precisely is more useful than collapsing every kind of risk into one absolute safe label.

Polymarket's resolution guidance emphasizes the governing rules and resolution process. For a proposed strategy, record which exceptional cases alter the expected payout or timing. If the analysis cannot establish how an exception affects the collection, represent that case as unresolved rather than assigning it the ordinary outcome by habit.

An operational failure should also remain distinct from a losing forecast. A mismatched token identifier, duplicated order or unreconciled fee is a process error. A directional residual losing after an incomplete hedge is a realized exposure. Classifying the result correctly helps a reviewer decide whether to change the contract filter, execution policy or accounting procedure.

Do not erase these distinctions after a favorable result. A mistaken one-sided position can make money by chance. That outcome does not validate the original arbitrage analysis. Review whether the account followed the intended coverage and execution rules even when the final balance increased.

Challenge the Claim With Boundary Cases

A practical proof exercise asks a reviewer to choose the least convenient permitted outcome. For a range collection, try the exact shared boundary, a value outside the displayed range and a missing official publication. For two related contracts, choose a path that separates their observation times. The purpose is to discover a zero-payout case before the account has paid for supposed coverage.

Then challenge the quantity assumption. Replace one intended leg with half its planned fill and calculate the remaining matched amount. This test often reveals that a tool is summing planned orders instead of actual holdings. Repeat it with a fee adjustment and a canceled remainder. A correctly structured record should change the economics without changing the underlying contract identity.

Finally, compare two explanations of a rejected candidate. One says the collection cannot establish the claimed payout. The other says the payout is established but acquiring it costs too much. Those are different research conclusions with different next steps. The first needs better contract reasoning; the second may need smaller size, different timing or a decision to leave the candidate alone. Preserving that distinction keeps a scanner's search for numerical gaps from replacing the more important task of proving what those gaps actually mean.

Use a Candidate Record That Can Be Challenged

A useful candidate record contains the identities, payout cases, quantities, executable cost estimate, hypothetical or current fee basis, completion path and invalidation conditions. It should be possible for another reviewer to disprove the candidate by identifying one unsupported assumption. That is more useful than an unexplained percentage labeled opportunity.

Include the reason for rejection when a candidate fails. A coverage gap, insufficient depth, excessive completion cost and unresolved order state are different findings. Keeping them separate shows whether the research process is mostly finding false relationships or valid relationships that are uneconomic to acquire.

The related Polymarket arbitrage software article describes how an evaluator can test that record in an application. Good reporting should show matched quantities and residual exposure beside the apparent gap. The most useful screen is the one that makes an incomplete claim difficult to mistake for completed profit.

Polymarket arbitrage analysis becomes clearer when the payout proof and acquisition plan remain separate. First establish exactly what the completed holdings return under the applicable terms. Then determine whether the account can acquire and complete those holdings at a favorable total cost. A candidate that fails either step is still useful research, but it should receive the correct explanation and remain outside the set of verified completed results.