A prediction market explained through one complete example is easier to follow than a list of disconnected definitions. Imagine a fictional contract about a reservoir's officially reported water level. Participants disagree about the outcome, post prices and trade claims that pay according to a precise rule. The event itself is imaginary, but the accounting questions are the same ones readers encounter in real markets.
The contract in this example pays one dollar per YES unit if a specified report shows the reservoir above 60% capacity on a named date. It pays zero on those YES units otherwise, assuming ordinary resolution and no exceptional provision. Every price, fee and forecast below is invented for education, not drawn from a real trading opportunity.
We will follow one researcher, Maya, as she reads the rule, forms a view, buys a smaller quantity than initially intended, sells part of it and reviews both possible final outcomes. The aim is to understand the process, not to suggest that Maya's forecast or strategy would make money in practice.
Prediction Market Explained With an Exact Rule
The phrase reservoir above 60% is not yet a complete contract. Maya identifies the reporting authority, the specific reservoir, the observation date, the units and the relevant publication. She checks that above means strictly greater than 60%, rather than at least 60%.
She also records whether the initial publication or a later corrected figure determines the result. A forecast based on the wrong reporting version would target a different event. This is a rule-reading problem before it is a probability problem.
The binary contract trading guide explains why the condition and payout belong together. A YES label has no useful financial meaning without the exact question that makes it pay.
For this fictional case, Maya's worksheet states that the named authority's initial daily report is controlling. A reported 60.0% means NO, while 60.1% means YES. Actual contracts can have additional details, but the example now has a condition that can be applied consistently.
The Forecast Comes From Evidence Available Before the Result
Maya examines historical reservoir observations and current conditions. She does not use later rainfall or the eventual published result. Her first estimate assigns a 58% chance that the contract's YES condition will be satisfied.
The number is uncertain. She writes a plausible range from 52% to 63% based on different interpretations of the available evidence. Recording that range helps prevent the central estimate from looking more authoritative than the research supports.
She notes which inputs are independent. Several reports repeating the same forecast do not count as several separate confirmations. She also states what new evidence would make her change the estimate, rather than waiting to invent a reason after the price moves.
A prediction market explained for first-time readers should preserve this uncertainty. The purpose of a probability estimate is to describe an uncertain event, not to give an opinion the appearance of mathematical certainty.
The Displayed Number Is Not the Entire Market
The screen shows a reference price of $0.50. Maya checks the order book and finds a best bid of $0.48 and an ask of $0.52. The midpoint is $0.50, but the displayed seller is asking more than the reference price.
She also sees that only 40 contracts are offered at $0.52, with another 100 offered at $0.55. This quantity information matters because her original idea was to buy 100. Applying the first ask to all 100 would understate the shown cost.
Official Polymarket price and order-book documentation illustrates the distinction between displayed references and execution prices. The fictional book here is not a quotation from that venue; it is a teaching device for the same general distinction.
The prediction market liquidity guide explains why depth changes the economics of size. Maya records both levels and the observation time rather than treating one price as available indefinitely for any quantity.
Maya Calculates the Proposed Purchase Before Clicking
If 100 contracts actually filled at the displayed asks, the first 40 would cost $20.80 and the next 60 would cost $33. Total purchase cost would be $53.80 before any fees. The average would be $0.538, not $0.52.
Maya adds an illustrative one-cent charge per acquired contract for this exercise. The all-in cost of that assumed full purchase would be $54.80. Under her central 58% forecast, expected payout would be $58 and expected net result $3.20.
At the lower end of her stated range, 52%, expected payout would be $52 and expected result minus $2.80. The same order can look favorable or unfavorable under plausible forecast assumptions. The worksheet makes that sensitivity visible.
She decides that the $0.55 level is beyond the price she wants to accept. This is not proof that the lower price is a good trade; it is a bounded instruction derived from her stated decision policy.
The Order Creates a Smaller Actual Position
Maya submits a buy instruction for up to 100 contracts with a $0.52 price ceiling and conditions that allow available quantity to fill while canceling the remainder. In this fictional sequence, the 40 contracts at $0.52 remain available and fill. The remaining 60 do not.
Her actual acquisition cost is $20.80 plus the invented $0.40 charge, or $21.20. She owns 40 contracts, not the original intended 100. The canceled remainder is a record of an uncompleted intention, not an additional position.
The difference is important for every later calculation. A chart based on 100 contracts would overstate both potential payout and loss. Maya saves the order identifier, fill quantity, actual price and cost rather than relying on the order-entry screen.
The article prediction markets explained places this example among different market designs. Here, the focus is narrower: only completed fills create the position being followed through the ledger.
The Two Ordinary Settlement Branches Are Now Clear
With 40 YES contracts and a $21.20 all-in cost, a favorable one-dollar payout would return $40. Ignoring any additional charges, net gain would be $18.80. An unfavorable zero payout would lose the $21.20 acquisition cost.
Under the central 58% estimate, expected payout is $23.20 and expected net result is $2. At 52%, expected payout is $20.80 and expected net result is minus $0.40. The limited purchase has changed the economics from the larger proposed order.
These expected values are not cash that Maya has earned. They are calculations based on her uncertain estimate. The actual event follows one outcome, and the account follows the position and costs that actually exist.
A useful journal keeps the two branches near the forecast. This prevents an attractive possible gain from being discussed without its associated loss and prevents the average of imagined outcomes from being mistaken for an available balance.
Prediction Market Explained Through a Trading Ledger
Maya's ledger begins with a $300 account balance. The purchase and illustrative charge reduce cash by $21.20 to $278.80. The account also holds 40 YES contracts. Cash and position value are separate entries, not two descriptions of the same amount.
Suppose a reference price later rises to $0.61. Multiplying it by 40 gives a $24.40 mark, but that is not a completed sale. If the current bid is $0.58 for the whole position, immediately realizable gross proceeds under an unchanged book would be $23.20 instead.
The mark-based gain is $3.20 relative to cost; the hypothetical immediate sale before exit charges would yield a $2 gain. Both calculations can be internally correct while answering different questions. The ledger should label which is shown.
The prediction market trading software guide discusses this distinction in product evaluation. A useful display lets Maya inspect the underlying prices and quantities rather than asking her to trust a single green profit number.
A Partial Sale Fixes Only Part of the Result
Maya later sells 15 contracts at an actual fill price of $0.62. Gross proceeds are $9.30. With an invented $0.15 exit charge, net proceeds are $9.15. Her cash balance becomes $287.95, and 25 contracts remain open.
Allocating the original all-in acquisition cost evenly gives a $0.53 basis per contract. The sold 15 have a $7.95 basis, so their realized gain is $1.20. The remaining 25 have a $13.25 basis that is still exposed to the event.
Selling part of the position does not mean the whole trade is now profitable under every possible outcome. If the remaining units become worthless, their loss can exceed the gain already realized. The final branches must be recalculated after the sale.
The guide to prediction market strategies explains why exit policy should be connected to the original thesis. Maya's reason for selling belongs in the journal alongside the actual transaction, rather than being inferred from the eventual outcome.
Refer & Earn USDC
Refer & Earn USDCOur affiliate program pays in real-time.
The YES Branch Produces One Final Account Result
Assume the controlling report shows 61.2%, satisfying the fictional above-60% condition. The remaining 25 YES units receive a one-dollar payout each, for $25. With no further charges in this example, final cash becomes $312.95 and the position closes.
Against the original $300 account, total gain is $12.95. The same result can be reconstructed from net sale proceeds of $9.15 plus settlement proceeds of $25 minus acquisition cost of $21.20. The two methods agree.
The sold 15 units are not included in Maya's settlement payout. Their subsequent economic outcome belongs to whoever held the relevant claims, subject to the market's actual mechanics. Counting them again would invent an extra $15 receipt.
This branch shows why a complete quantity history matters. A correct final event forecast does not by itself establish the account's profit; the entry size, partial sale and costs determine the actual financial result.
The NO Branch Produces a Different Final Result
Now consider the alternative in which the controlling report shows 59.8%. The remaining 25 YES units pay zero under the fictional ordinary rules. Maya keeps the $287.95 cash balance after the partial sale and has no remaining position value.
Relative to the starting $300, the final loss is $12.05. It can also be calculated as the $21.20 acquisition cost minus the $9.15 net sale proceeds. The partial exit reduced the loss compared with holding all 40 units to a zero payout.
That does not prove the exit was optimal or that early selling is generally preferable. In the YES branch, selling the 15 units early also gave up their later one-dollar payout in exchange for the earlier proceeds. The decision traded one payoff profile for another.
A prediction market explained with an example should include both branches. Showing only the favorable final report would turn an educational ledger into a misleading story about an apparently inevitable gain.
A Boundary Result Tests Whether the Rule Was Understood
If the initial report shows exactly 60.0%, the fictional strictly-above condition produces NO. Maya cannot replace that rule with at-least-60% because the outcome feels close. The difference was part of the claim at purchase.
Suppose the authority later corrects the report to 60.2%. In this teaching case, the rule uses the initial publication, so the correction does not automatically change the ordinary determination. A real contract's wording and exceptional provisions would need to be checked rather than assumed.
Official Kalshi market guidance is another example of why market-specific terms matter. The fictional case is not a claim that every venue handles corrections in the same way.
This is a useful exercise before discussing profitable strategies. If the reader cannot apply the boundary case consistently, the probability estimate may be aimed at a different condition from the contract being traded.
Market Probability Is a Signal, Not a Verdict
Maya's 58% estimate and the market's prices can disagree without either being proven wrong immediately. The event has not yet resolved, and both reflect uncertainty. The useful question is what evidence supports each view and what execution terms are available.
The article on how prediction market probabilities work explains why a price is not an objective probability certificate. Costs, risk preferences, information differences and liquidity affect what people are willing to trade.
A later rise in price can reflect new evidence, changing demand or a thinner book. Maya should record the observation separately from the explanation. A convincing narrative about a chart does not substitute for evidence that the narrative caused the movement.
Likewise, one favorable final result does not validate the 58% estimate. Calibration is examined across suitable groups of forecasts, with original timestamps preserved. A single binary outcome cannot establish the accuracy of a precise probability.
The Example Changes When Execution Is Interrupted
Imagine Maya's application loses connection after submission. She sees no confirmation and is tempted to click again. If the first order already filled, a second independent instruction could create another position. The missing response does not prove the first attempt failed.
The recovery task is to inspect the original order and fill records before creating fresh exposure. If the account shows an uncertain remainder, that state should be reconciled. A clean-looking screen obtained by ignoring the uncertainty would be less useful than an explicit pending investigation.
A similar problem occurs with cancellation. A request to cancel can arrive after part of the order has matched. The canceled remainder and the acquired position need separate records, even when the user intended to avoid the entire trade.
These operational branches are not distractions from the financial example. They determine whether the ledger describes the account that actually exists. Understanding them is part of understanding a prediction market.
Scaling the Example Does Not Preserve the Same Economics
Multiplying Maya's 40-contract result by 100 would not establish what a 4,000-contract trade could earn. The displayed depth, available capital and price impact could be very different. The larger instruction might consume several price levels or fail to complete.
The original forecast uncertainty also remains. Increasing quantity increases the dollar consequences of a wrong estimate without improving the estimate itself. A small positive modeled expectation is not a reason to assume unlimited capacity.
If other contracts in the account depend on the same reservoir conditions, total exposure may be concentrated. Several apparently distinct positions can share one adverse weather scenario. The risk review should examine those joint outcomes rather than each order in isolation.
This is why a worked example should be used to learn relationships, not to project income. The arithmetic is reproducible under stated assumptions; the availability and reliability of those assumptions require separate evidence.
External Funding Does Not Change the Trade's Profit
Suppose Maya deposits another $100 after the partial sale but before settlement. Under the favorable branch, final cash would become $412.95 rather than $312.95. That larger balance would not turn the trade's $12.95 gain into a $112.95 gain; the extra $100 came from outside the strategy.
If she instead withdraws $50 during the same period, the favorable ending balance becomes $262.95, with $50 already returned to her. Adding the withdrawal back before comparing with the original $300 still gives a $12.95 trading result. Cash movement and investment performance answer different questions.
This variation is useful because informal screenshots often show only a balance or a withdrawal. Neither can establish profit without the starting capital, external flows and remaining positions. Maya's event ledger remains the same even when she moves unrelated funds through the account.
For a complete review, preserve the date and purpose of every funding movement separately from order fills. Then reconcile the whole account using a consistent valuation time. The result should agree with the sum of completed position records and any explicitly valued open exposure, rather than relying on a favorable-looking number selected from one screen.
Use the Case as a Review Exercise
Try changing one input while preserving all others. Increase the acquisition charge, reduce the actual fill quantity or move the partial-sale price. Recalculate the YES and NO branches and confirm that the account balance and transaction ledger still agree.
Then change a rule rather than a price. Replace strictly above with at least, or change the controlling source version. Notice that this alters the event being forecast, not merely the accounting. It may require a new probability estimate before any trade calculation is meaningful.
Finally, ask what information was missing at each stage. The first screen did not show the full cost of the intended quantity. The mark did not establish liquidation proceeds. The apparent result did not replace settlement verification. Recognizing those gaps is the main educational value of following one complete case.
A prediction market connects a precise question to tradable claims and financial records. Once those connections are visible, a reader can investigate real examples more carefully, challenge confusing displays and decide when the available evidence is insufficient to participate.