Chronomancy

The Problem

Prediction markets have a retention crisis. 70% of trading addresses lose money. Median retail ROI is −27%. Post-election MAU dropped 57%. The top 1% of users generate ~70% of all volume. Everyone else churns.

Traditional gambling softens losses with comps, bonuses, and entertainment value. Prediction markets offer nothing — you lose your money and feel stupid. No skill progression, no safety net, no reason to come back.

The Solution

Chronomancy wraps existing prediction market platforms — Polymarket, Azuro, Kalshi — with structured financial products that make forecasting viable at scale.

Insurance

Recover up to 70% of losing positions via REWIND — actuarially priced with segregated risk pools, concentration limits, and CS-based discount tiers. Not a subsidy; a real insurance product where premiums fund claims.

Early Exit

Cash out positions before resolution via FAST-FORWARD. The vault pays above market price for high-CS forecasters because their positions resolve correctly more often — this is alpha harvesting, not charity.

Reputation

Build a soulbound forecasting reputation (Chrono Score) based on prediction accuracy — difficulty-weighted, Bayesian-shrunk, time-decayed. Better scores unlock better financial terms.

Seasons

Quarterly LOOP epochs reset competitive rankings and leaderboards. Your capital and positions are untouched — reputation persists, competition refreshes.

How It Works

Chronomancy doesn’t replace prediction markets — it sits on top of them. You trade on Polymarket or Azuro as usual. Chronomancy reads your positions, computes your Chrono Score, and lets you activate modules (insurance, early exit, etc.) on those positions.

→ See the platform architecture

”Isn’t This Just Subsidizing Losers?”

No. Chronomancy is not a zero-sum intervention. Insurance premiums are real revenue — users pay upfront for coverage, and the protocol is profitable on aggregate (premiums > claims). The insight is about retention economics: traditional prediction markets permanently lose 70% of users after their first loss. If insurance retains even 30% of those users for one more prediction cycle, the protocol earns more in total fees than it pays in claims. The retained user generates FF spread revenue, staking activity, and CS data that compounds over time. Rewind doesn’t fix the math of individual bets — it fixes the unit economics of user lifetime value.