Bot Titles
POL-BTC-HEDGE-V1Performance So Far
POL-BTC-HEDGE-V1Recent Fire Attempts
POL-BTC-HEDGE-V1Token Reverters (24h)
POL-BTC-HEDGE-V1Biggest Block
POL-BTC-HEDGE-V1The edge is mathematically sound but operationally UNPROVEN — it never actually traded. The math is airtight: buy YES+NO for a combined ≤ $0.95 and one side resolves at $1 = risk-free ~5%. But the history (2026-03-09 → 04-16, 2,104 log lines) shows 0 trades ever executed — it ran mostly in paper mode, repeatedly logged "Startup: found 0 BTC markets," and the entire recent period is wall-to-wall network-timeout errors against its data feed. So the real blocker is NOT the strategy — the bot could never reliably find the markets or read the books: its data layer (a homemade Cloudflare-Worker proxy, polymarket-proxy.trevorgdhawg.workers.dev) kept timing out and the BTC-market discovery filter returned nothing. The edge's real-world frequency + fillable depth were never even observed.
Fantasy
POL-BTC-HEDGE-V1If we had some way to instantly remove this bot's biggest constraint — capital, latency, signal coverage, or counterparty risk — what would the world look like? (Pending creative assessment.)
Suggestions for Improvement
POL-BTC-HEDGE-V1- 01CRITICALPhase 0 — rebuild the data layer. The bot never traded because its homemade Cloudflare-Worker proxy (polymarket-proxy.trevorgdhawg.workers.dev/sampling-markets) kept timing out and the discovery filter returned "0 BTC markets". Replace with Gamma API (gamma-api.polymarket.com) for market discovery + the CLOB API for order books, and fix the BTC 15-min market match.
- 02CRITICALPhase 1 — observe-only validation (3-7 days, NO capital). The make-or-break. Log every BTC 15-min round YES+NO sum, every >=15% dislocation, and the fillable book depth below $0.95. Go-live ONLY if the edge appears often enough with enough depth to clear fees + slippage — the strategy was never measured live (0 trades ever).
- 03HIGHPhase 2 — harden the CLOB order path before any live capital. Fix the order-signing nonce handling (the shared CLOB-exploit risk) and make both legs fill-or-kill with an immediate unwind if only one leg fills — a one-sided fill is naked directional BTC exposure, the opposite of a hedge.
Conclusion
POL-BTC-HEDGE-V1Verdict (2026-05-26): the most revivable paused POL bot — but VALIDATE-FIRST, not resume. This is the only paused Polygon bot with a structurally sound, risk-free edge (dual-leg ≤ $0.95 → guaranteed $1). But pulling its history was sobering: it executed 0 trades across its entire 2026-03-09 → 04-16 run (mostly paper, found 0 BTC markets, then died on data-feed timeouts), so the edge has never been captured or even measured. The revival is therefore validate-first, not a restart.
Plan. Phase 0 — data layer: replace the flaky Cloudflare-Worker proxy with the Gamma API (market discovery) + CLOB API (order books); fix the filter that returned 0 BTC markets. Phase 1 — observe-only (3-7d, the make-or-break): log every round's YES+NO sum, ≥15% dislocations, and fillable depth at sub-$0.95; proceed ONLY if the edge appears often enough with real depth to clear fees. Phase 2 — harden the CLOB path: fix the order-signing nonce (the shared CLOB-exploit risk) + make both legs fill-or-kill with an unwind path (a one-sided fill = naked directional exposure). Phase 3 — paper → small live: $46 balance, $20/window, circuit breaker + daily loss cap.
Honest read: Phase 1 is decisive — a cheap test of whether the edge is real before any capital or build. Even if it validates, 15-min BTC liquidity caps it at a small steady carry, not a flagship earner.