Measured to the date+time of the winner's announcement by TIME.
Days = 24 hours
Constitutes as leaking if any of the following happen:
-Clear, sustained market movement on Polymarket or Kalshi or even Manifold (especially if it's an option not listed on PM/Kalshi) indicating that the winner is known (the eventual winning option goes >80% for more than a transient spike)
-The leak is published in a newspaper / on a blog / etc with a well-described rationale of how they achieved the leak, and this is validated by later reporting.
-Other things that are really obviously leaks and are immediately apparent to people as a leak, that I haven't thought of.
I will not bet on this market. I'll extend the date until the announcement date/time is apparent or occurs.
Update 2025-11-30 (PST) (AI summary of creator comment): In edge cases where the timing of a leak is ambiguous (e.g., gradual market movements vs. sudden spikes), the creator may:
Consult analysis from Polymarket commentators and other expert sources
Resolve partially to different options if the leak timing is unclear
1,000People are also trading
@bens I like this market, but I want to ask something important before I commit:
Since there is some discretion in determining exactly "if it leaked, and if so, when", I want to hear about how you plan to assess the situation in the event of any question. Is this something you intend to decipher and decide unilaterally? Am I betting on "How will Ben S read the situation?" or am I betting on "How will the oracles Ben S decides to consult read the situation?"
Hope that makes sense.
@Eliza hmmm, ya, I think it'll probably be something like "How Ben S will read the situation, with Ben S having pretty good humility and trusting other users, consensus sources, common-sense interpretations of the news, etc"
I think the biggest edge cases would be stuff like:
Jensen Huang is at 40% the week before... 4 days before, it goes up to 55%, then 2 days before, it goes up to 85%. Do I think the leak happened 4 days before or 2 days before.
I think in cases like these I may resort to:
-trusting the analysis of, for example, Polymarket twitter people who write analyses of stuff like this
-reserving the ability to resolve partially to different options
@bens Alright, that sounds good. I just wanted to double check this wasn't going to be "Resolves to a random poll". I trust you!