Simulation, not investment advice. Forward-only copies use delayed public signals and public CLOB book snapshots; modeled fills are not executable quotes. Results cannot establish profitability.
FEED: CONNECTINGUPDATED: —BASELINE: —
Equity curve
Source wallets
| Rank | Wallet | Board PnL | Exposure |
|---|
Open paper positions
| Market / outcome | Shares | Avg | Mark | Mark age |
|---|
Recent paper fills
Signal audit
| Observed | Market token | Decision | Reason |
|---|
Model boundaries
Wallets are selected from the current monthly PnL leaderboard, creating survivorship and ranking bias. Finite-window rankings include mark changes; they are not verified skill. The forward-only baseline cutoff is the later of poll start and latest source timestamp, so earlier trades cannot be filled if they arrive late. Too-recent signals remain pending until the modeled delay matures. Each observed token/book hash can back at most one modeled fill; later reuse is rejected. Fills walk the observed public book and apply the published fee curve, but network latency, queue position, book changes, rebates and rounding can differ from execution.