open‑snipe

A brand-new market has exactly one price in it, and that price is whatever its creator picked. Markets of the same shape — same topic, same opening price, same horizon, same ante — have a resolution rate the venue has already revealed. When that history and the creator's price disagree by more than 8%, this module bets 50% of the way from the price toward the history. It only works on markets nobody has traded yet, so its real competitor is not a price — it is everyone else who got there first. It is running in shadow in this version: every bet below is paper, judged like a real one but never sent.

The race for untraded markets

Every market in the second column cleared this module's rules and was lost to whoever arrived sooner. That is the only bucket here that represents money left on the table. Markets already traded when we first looked never reached the rules at all, so they are counted apart; the rest are markets it did not want.

What the history said, market by market

Dark tick is the creator's opening price; coloured tick is what the history said — on a decline, the rate markets of this shape resolve at; on a bet, the price this module moved it toward. The bar between them is the edge, and its colour is the side of the bet — warm toward YES, cool toward NO. Rows marked bet are the journal's; the rest are traces.

Where candidates died

The book · v86

43 trades, 867 staked, realised -257, open +18 if held / -6 if sold, 2 unfilled.

Of 0 markets judged in the traces on hand, 43 cleared every rule and 43 were reached before anyone else, for 867 mana. Of those, 6 settled for -257. Losing a market to a faster participant is not a rejection and is counted separately from one.
Traces are diagnostic and rotate; the trading journal is the book above. Built 2026-09-22 18:01Z.