Last Updated: 2026-10-25
Risk: This is a selection worksheet with hypothetical arithmetic, not a list of current trades or a return forecast. A binary position can lose its whole stake.
A useful shortlist starts with questions you can resolve and evidence you can inspect. “Politics,” “sports” or “crypto” is not itself an edge. This guide replaces undated market picks and unverified personal results with a record that remains usable when quotes change.
Screen the contract, the information and the exit separately
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Market type | Evidence to collect | A reason to reject the idea |
|---|---|---|
| Scheduled economic announcement | Exact release, source, revision treatment, cutoff and timezone | A binary announcement is being compared with a monthly-average futures payoff |
| Sports event | Exact competition, cancellation/overtime rules and timestamped team news | The quoted price predates an injury update, or the settlement rules differ from the comparison venue |
| Election or legal outcome | Exact event, authoritative result and recount/appeal treatment | A related state, party or court event is being treated as the same contract |
| Crypto price threshold | Exchange/index, observation interval, touch versus closing-price rule | A path-dependent “touch” event is compared with an option paying only at expiry |
| Long-dated novelty event | Verifiable source, ambiguity handling and plausible exit depth | The thesis relies on being able to sell quickly into an empty book |
These are research distinctions, not recommendations for a category. Use the provider's resolution documentation and the particular market rules. A headline is not enough.
Inspect the quantity you would need to execute
Save bid/ask levels and quantities with a timestamp. Historical volume can be large while current depth is small. If the displayed ask offers 100 shares at $0.40 and the next 100 at $0.45, buying 200 at those levels costs $85, averaging $0.425 before fees—not $80. If only the first level fills, your position and remaining order are different from the planned 200-share trade.
Do the same for a possible exit. Mark a hypothetical result as a simulation; an order-book snapshot cannot prove a later fill. Check current fees for the market, platform eligibility and whether funds can remain committed through delayed resolution.
Use the initial outlay as the return denominator
For an illustrative bearish position, buying NO at $0.82 and later selling at $0.96 gains $0.14 per share, or 17.07% before costs. Dividing by the later YES price of $0.04 would produce a meaningless 350% figure for that position.
Buying YES at $0.92 and receiving $1 produces 8.70% gross profit on stake. Receiving zero loses 100% of the $0.92 stake, not 92%. A high quoted probability does not make the downside small relative to invested money.
For another synthetic example, $6,000 buys 6,000 / 0.38 shares. Selling half at $0.49 and settling half at $1 returns $11,763.16, for $5,763.16 profit before costs. This arithmetic is not evidence that a corresponding historical trade happened.
Time and hedges add conditions, not certainty
Compare the expected net result, plausible delay and capital committed for each idea. A small gain locked up for six months is different from the same gain over a week, but annualizing a one-week success assumes repetition that has not been demonstrated. A longer horizon does not justify accepting less evidence simply because the deadline is distant.
Write the payoff under every relevant joint outcome before calling two positions a hedge. A rate future based on a monthly average does not exactly offset a binary meeting decision. Related election contracts can settle differently. See the payoff-table guide for the distinction between a complete basket and correlated exposure.
A shortlist row that can be challenged
Record: market URL and ID; review time; rules and source; evidence for and against; probability range or “unknown”; intended quantity; entry and exit depth; all-in cost assumptions; maximum loss; potential delay; and a clear reason to stop considering the idea. Reject an incomplete row instead of filling it with a guessed price, “typical edge” or recommended allocation.
Keep rejected ideas as well as selected ones if you later evaluate the process. Otherwise the review becomes a collection of memorable winners. AITR may receive commissions from marked referral links elsewhere; no current market or platform signup is recommended on this page.