Read the Fine Print: How to Check a Take Before You Call It

Here is an uncomfortable pattern we see constantly. A player has genuinely good knowledge — they follow the sport, they know the release calendar, they were right about what would happen. And they still lose the call, because the take was not asking the question they answered.

This is the cheapest mistake to fix on the entire platform. It costs you five seconds of reading per market. This article is the checklist.

1. Find the verb, and find the threshold

Start by reducing the question to its skeleton: who, does what, by when, measured how.

"Will Team A win the series?" — what counts as winning a drawn series? Does the holder retain it?

"Will the film cross ₹100 crore?" — worldwide or domestic? Gross or net? In how many days? In which currency, if it's a global release?

"Will the artist release an album this year?" — does an EP count? A deluxe reissue of an existing record? A soundtrack they contributed to?

A well-written take answers all of that in its criteria. Your job is to confirm it does, and to stop and think if it doesn't. If a take genuinely leaves the answer ambiguous, that ambiguity is a risk you are taking on, and the fair outcome for a truly ambiguous question is a void — which returns your Stars but also wastes the days you had them committed.

2. Check the window, not just the deadline

Two dates matter and they are not the same.

The acceptance deadline is the last moment you can make a call. The resolution deadline is when the question is due to be answerable.

The gap between them is where most confusion lives. A take can stop accepting predictions well before the event — that is deliberate for anything where information arrives in a rush near the start, like a team announcement or a first-day figure. If you were planning to wait for the team sheet, check that the market will still be open when it lands.

Then look at the resolution deadline and ask: is that actually enough time for the answer to exist? Some answers are instant. A match ends and a scorecard exists. Others take days: certified box-office figures, official confirmations, governing-body rulings. If the resolution deadline is tight against a source that is historically slow, the take is more likely to need a recheck cycle — your Stars are committed for longer than you planned.

3. Identify the source of truth

Every take should be answerable against a specific, checkable class of source. Before you call it, know which one it is:

• A scorecard or official result. Fastest and least ambiguous. If the take rests on one of these, it will almost certainly settle cleanly and quickly.
• An official announcement. Reliable, but on the announcer's schedule, not yours. "Will X be announced by date Y" is fundamentally a question about someone else's PR calendar, and PR calendars slip.
• A reported figure. Box office, streaming charts, viewership. Check which reporter the take names, because different trackers routinely publish different numbers for the same film on the same day. A take that says "as reported by [source]" is a much better take than one that says "reportedly".
• A subjective or contested judgement. Be careful. If two reasonable people reading the same evidence would disagree about the answer, so will the resolution process.

The single best habit you can build: prefer takes with a clean source. Not because the others are unfair, but because they carry a different kind of risk — a risk about whether the answer will be legible, on top of the risk about what the answer is.

4. Hunt for the sneaky word

A handful of words do enormous work in a prediction question. When you see one, slow down.

"Officially" — narrows the source to a specific authority and rules out credible reporting. A thing can be true, widely reported, and still not officially confirmed by the deadline.

"Consecutive" — a much higher bar than the same number of events spread out.

"Or more" / "at least" / "over" — check whether the threshold value itself counts. "Over 100" and "at least 100" resolve differently on exactly 100.

"Any" — usually broader than a first read suggests. "Any player" includes the ones you weren't thinking about.

"Announced" vs "released" vs "confirmed" — three different events, often months apart, routinely treated as interchangeable by readers and never by resolvers.

"During" — bounded by the window. A thing that happens the day after is a No, no matter how close it came.

5. Ask what makes this a Void

Run the question through the obvious disruptions and ask what the criteria say happens:

• The match is abandoned or rained off.
• The film's release is postponed past the window.
• The event is cancelled.
• The player is injured before they participate.
• The source stops publishing the figure.

A carefully written take answers these. If it doesn't, and the disruption is plausible for this question, factor that in. Rain in a specific city in a specific week is not a footnote — for some markets it is a meaningful share of the probability mass, and a market priced as if the game will certainly be played is a market with a hole in it.

This cuts both ways, and it's one of the more reliable edges available to a careful reader. If a take voids on abandonment and there is a real chance of abandonment, then the effective question is "given the event happens, will X?" — which may be a very different probability from the one the price implies.

6. Re-read it once your position is large

Everything above applies double when you are about to commit a lot of Stars. Our internal rule of thumb, and we recommend adopting it: when you find a market that looks obviously mispriced by 30 points or more, the first hypothesis is that you have misread the question. Sometimes you have genuinely found something. Far more often, the word "officially" is sitting in the criteria doing quiet work.

Read it again. Then decide.

Why we write it this way

Everything in this checklist is on the take page before you commit anything — the criteria, both deadlines, the source class. That is deliberate. A market that hides its terms until after settlement is not a prediction market, it's a trap with a leaderboard attached.

We also do not edit resolution criteria after a take opens. The question you read when you made the call is the question that gets answered. If we discover a question was written badly, the fix is to void it and write a better one, not to quietly redefine what people were playing.

Read the fine print. It's the only part of forecasting where the effort is trivial and the return is guaranteed.