Decentralised News · Agentic Finance · Decision research
By Heath Muchena · Source review 2026-11-14 · Instrument version 1.0.0
What Matters
A prediction-market price can express a market view, but a purchase decision also depends on exact resolution rules, executable quantity, fees and capital timing. DN’s record calculates the event probability needed to recover entered costs in a confirmed two-outcome model. It does not estimate the true probability or recommend a trade.
A contract quoted at 70 cents may be described as a 70% market. That shorthand can be useful, but it does not answer whether a particular buyer would recover the purchase cost after fees or whether the contract resolves the event the agent has in mind.
This article separates the forecasting question from the contract-economics question. The original DN instrument records payout assumptions and computes a conditional cost-recovery threshold. The numerical examples are synthetic; no market, trading strategy or forecasting advantage was tested.
DN Evidence Block
Sources reviewed: 2026-11-14 · Evidence type: Primary Kalshi documentation and an original two-outcome cash-flow method; sources checked on 13 November 2026 UTC.
Scope: One fully paid position held to settlement, one currency, two documented payout states, user-entered costs. Other payout structures remain outside the formula.
- Primary market documentation distinguishes contract rules, fees, payout structure and determination timing.
- A price can reflect market expectations without equaling the buyer’s after-cost recovery threshold.
- Synthetic arithmetic illustrates a 70-unit purchase plus 2 units of fees requiring a 72% threshold under a 100-or-zero payout.
Not established by this work: No live quote, fee estimate, execution, settlement, probability forecast, calibration result or trading outcome was observed. The tool does not determine legal access or suitability.
Original DN instrument · version 1.0.0
DN Contract Cost Recovery Record
Calculate conditional cash and alternative-cost recovery thresholds for one documented two-outcome position. Unknown fees or payout scope remain unresolved. No market probability, trade recommendation or live price is produced.
Enter your assumptions or load a labelled example. Required blank numbers stay missing; unknown evidence remains unknown. Use nonpersonal labels only. This instrument makes no network requests and uses no persistent browser storage.
Use your own inputs
No current result. Complete inputs or load an illustrative example.
A user-entered conditional model, not a forecast. Assumes one currency, a fully paid position, whole contracts and exactly two settlement payout states held to settlement. Excludes leverage, early exit, variable quantity, tax, FX, disputed settlement and portfolio correlation. Fees and rules are not fetched or verified. Holding days are context only; alternative cost is entered separately. No account or network access.
1. Read the exact event before interpreting the percentage
A price can be informative about what market participants expect. It is still a price for a contract with a particular outcome definition. Before an agent turns that price into a probability, it needs to know what must happen, which date or threshold applies, and what the settlement authority will recognize.
Kalshi describes contract prices as reflecting traders’ views of event chances. Its market-rules guidance directs users to the criteria and verification source for each market. Those two statements belong together: the market view concerns the defined contract, not every ordinary-language interpretation of the headline.
Record: the exact contract identifier, outcome wording, cutoff and governing source. Two markets with similar titles can differ in what counts as a successful outcome. This article does not claim that prediction prices are never probabilities. It shows which additional assumptions are needed before a quoted percentage can inform an economic decision.
2. Establish the payout shape before using a binary formula
The familiar one-or-zero example is useful only when the contract actually has those two settlement payoffs. A product with a fair-value adjustment, proportional payout or another exceptional outcome needs a different model.
Kalshi’s portfolio guidance illustrates a simple binary contract paying one unit on a favorable outcome and zero otherwise. Its ladder and escalator documentation describes products with different payout mechanics. The market’s terms govern; the product category alone does not supply a universal payoff.
Record: the winning and losing payouts and whether additional settlement states are possible. DN’s instrument withholds its binary threshold when the structure is unknown or has other outcomes. Entering a familiar one and zero does not override an unresolved contract rule. A model with missing states can produce precise arithmetic for the wrong financial instrument.
3. Use the price available for the intended quantity
A headline price, the latest trade and the price available for the proposed order are different observations. The decision needs the actual entry-price assumption and the number of contracts it applies to. Record whether the amount is a quote, an executed average or a synthetic scenario.
Kalshi’s pricing explanation describes matching buyers and sellers. That mechanism does not promise that an agent can buy any quantity at the last displayed percentage.
Record: quantity, entry price, observation time and whether the intended order was filled. If several fills have different prices, use the documented quantity-weighted average and retain the fills behind it. An unfilled order is not a position. The local tool performs no order-book lookup and does not estimate slippage or execution probability.
4. Add the applicable fees without assuming a universal schedule
The quoted percentage is not automatically the cash break-even point. Entry fees increase the amount that must be recovered; outcome-specific charges can also change the difference between winning and losing receipts.
Kalshi’s fee guidance says some markets use different fees and some resting orders incur maker fees when executed. It links the operative schedule. That supports checking the exact market and transaction rather than hardcoding a fee rule across products.
Record: total entry fees and any additional costs conditional on each settlement outcome. A blank is unknown, not zero. Do not add an exit fee to a hold-to-settlement scenario unless that fee actually applies. Likewise, do not call the result after-fee if the fee source or side of the trade is unresolved. DN’s tool requires explicit fee inputs and an evidence reference before displaying a threshold.
5. Calculate the probability needed to recover the entered cash
For a two-outcome contract, the economic threshold comes from the entered cash flows. Let C be initial cash outlay, W the net receipt if the winning outcome settles, and L the net receipt if the losing outcome settles. If W is greater than L, the probability p that makes expected cash receipt equal outlay is p = (C − L) / (W − L).
Synthetic example: one hundred contracts at 0.70 currency units each plus two units of entry fees cost 72. If they pay 100 in the winning state and zero in the losing state, the cash cost-recovery threshold is 72%, not the quoted 70%. Winning would produce a net cash result of 28; losing would produce −72.
This is a consequence of the stated model, not a forecast. It does not establish that the true event probability is above or below 72%. If the threshold exceeds 100%, none of the possible probabilities in that two-outcome model recovers the entered outlay. The tool preserves that result rather than clipping it to an apparently plausible percentage.
6. Separate the event date from when capital becomes available
An event can appear finished while the contract awaits its governing confirmation. An agent should not spend the anticipated payout as if it were already available cash.
Kalshi’s timeline guidance distinguishes markets that close when an outcome occurs from those that close at a listed time. Its settlement FAQ further distinguishes market close from determination and notes that official data can delay settlement.
Record: the expected holding period, the source for that expectation and the consequence if funds remain committed longer. DN’s tool accepts a holding-period assumption as context and an optional explicit alternative capital-use cost. It does not turn a date into a guaranteed payout or apply an assumed interest rate. If the alternative use is worth three currency units in the synthetic example, recovering the original 72 plus that scenario cost requires 75% under the same payout model.
7. Treat settlement-source and exceptional-outcome risk explicitly
The ordinary-language event and the contract’s official determination can diverge because the contract uses a particular source, revision policy or timing rule. The relevant question is what the contract resolves to, not what the agent believes happened in the world.
Kalshi’s outcome guidance places the determination source and procedure in the contract terms. Its market FAQ describes cases where some products can settle at a nonstandard value. These are reasons to inspect the actual rule, not evidence that any specific market will resolve incorrectly.
Record: ambiguous wording, possible revisions, exceptional payouts and the unresolved review question. If a third payout state is material, use a model with that state and its assumptions. Do not hide resolution uncertainty inside a guessed probability adjustment. DN’s two-state tool deliberately refuses to summarize such a contract as though it had only two outcomes.
8. Keep forecast accuracy separate from contract economics
A forecaster can be well calibrated and still face an unattractive purchase price after costs. A contract can have a low cost-recovery threshold without giving the buyer evidence that its event estimate is accurate. These are different assessments and should remain separate in the agent’s decision record.
The economic worksheet does not need a prediction of the true probability to show the threshold. That design prevents a model-generated confidence number from quietly becoming the justification for a trade. A separate forecasting evaluation needs dated predictions, observed outcomes and a suitable scoring method.
Record: the economic assumptions here and the forecast evidence elsewhere. Keep the original quoted price alongside any cost-adjusted threshold, so a later reader can tell whether a changed number came from market information, fee assumptions, quantity or a different contract definition. Neither side alone establishes that a trade should be made.
9. Export the assumptions, not an unexplained percentage
An agent consuming a machine-readable result needs the contract scope and arithmetic inputs, not just a final percentage. Keep the rule and fee references, currency, quantity, payout structure, missing values and creation timestamp with the calculation.
Record: which values were observed and which were hypothetical. The worksheet’s examples are synthetic and remain labeled that way in the local export. It does not fetch market data, connect an account, place an order or verify the documents supplied by the user.
A reviewer should be able to recreate the result from the same inputs. If a fee changes or the contract is found to permit another settlement value, create a revised record instead of silently changing the old number. The useful output is a conditional economic model whose limits are visible, not an assertion that the market’s displayed probability is false.
Methodology, limitations and corrections
For quantity N, entry price q and entry fees f, initial outlay C = Nq + f. Net winning receipt W = N × winning payout − winning-state costs. Net losing receipt L = N × losing payout − losing-state costs. The expected receipt in the model is pW + (1 − p)L. Setting that equal to C yields p = (C − L)/(W − L), provided W exceeds L. An explicit alternative-use amount A changes the hurdle to (C + A − L)/(W − L).
The displayed threshold is not clipped to 0–100%. A value above 100% exposes an unrecoverable entered outlay in the model; a negative value exposes unusual assumptions under which both modeled states cover it. Unknown cost inputs remain null. Unknown or nonbinary payout scope, missing contract/fee references, or a nonpositive net payout difference withholds the probability threshold. The user must verify all cited rules and values independently.
The synthetic reference case uses N = 100, q = 0.70, f = 2, payouts of 1 and 0, and no further outcome fees. It produces outlay 72, net outcomes 28 and −72, and a 72% cash threshold. Adding an illustrative alternative-use amount of 3 produces 75%. These calculations are reproducible properties of the example, not evidence about a market’s true odds.
The local JSON export preserves the contract and fee references, entered assumptions, result and creation timestamp. No request is sent to a trading platform. Version 1.0.0, 14 November 2026: original cash-flow instrument and synthetic validation fixtures, with primary sources checked on 13 November UTC.
For corrections, identify the claim, source and date when contacting Decentralised News. A new record-creation timestamp does not refresh its underlying evidence.
Change log: version 1.0.0, 2026-11-14. Initial documentation review, original instrument and labelled examples.
Frequently asked questions
Is a 70-cent contract never a 70% probability?
That is too broad. A normalized binary price can be interpreted as a market view under relevant assumptions. It is not automatically the buyer’s after-cost recovery threshold or a verified estimate of the true event probability.
What does the 72% result mean in the example?
Under the entered two-state payouts and costs, a 72% winning probability would make expected cash receipt equal the 72-unit outlay. The tool does not establish that the event has that probability.
Why not set missing fees to zero?
Zero is a substantive assumption. Missing fees can materially change the threshold, so the tool leaves it unavailable until the relevant costs are entered explicitly.
Does the holding period automatically add interest?
No. Days are contextual. Any alternative capital-use cost must be entered separately and remains a scenario assumption, not a promised return.
Can the tool handle a fair-value or third settlement outcome?
No. Mark that structure as other or unresolved. A model with more payout states and their assumptions is needed.
Does a threshold below my forecast tell me to buy?
No. Forecast quality, execution, settlement, eligibility, risk and alternative uses remain separate questions. This record provides conditional arithmetic only.
Primary sources
- Kalshi: what are prediction markets?. Contract prices as participant expectations and simple winning payout framing. Reviewed 2026-11-13.
- Kalshi: market rules. Contract-specific resolution criteria and verification sources. Reviewed 2026-11-13.
- Kalshi: how prices are determined. Buyer/seller matching and price interpretation; no adopted guarantee of accurate forecasts. Reviewed 2026-11-13.
- Kalshi: fees. Market-specific and maker-fee variation; operative fee schedule must be checked. Reviewed 2026-11-13.
- Kalshi: timeline and payout. Market closure timing differs by contract. Reviewed 2026-11-13.
- Kalshi: market outcomes. Determination governed by contract criteria and source data. Reviewed 2026-11-13.
- Kalshi: market FAQs. Determination versus close time, official-data delays and some nonstandard settlement values. Reviewed 2026-11-13.
- Kalshi: portfolio balances and positions. Payout versus profit and a simple one-or-zero contract example. Reviewed 2026-11-13.
- Kalshi: ladder and escalator markets. Existence of payout structures outside the simple binary formula. Reviewed 2026-11-13.
Turn the finding into a decision
For a team comparing contract definitions or documenting payout assumptions, the DN Research Desk can discuss a scoped, source-based research brief. Agree the contracts, access and deliverable first. No event probability, profitable position or legal eligibility is promised, and no trading-platform referral is needed to use this record.
Related DN research
Disclosure: Analysis and instrument results are editorial. Where labelled affiliate links appear, DN may receive compensation when you sign up or buy through them. Commercial relationships do not determine findings. No tested integration or guaranteed outcome is implied.
