Polymarket Copy Trade Poly Syncer Debuts Non-Custodial Mirroring Engine

A managed execution layer for prediction markets pairs a statistically filtered wallet leaderboard, sub-three-second order replication, EIP-712 scoped permissions, and a Trail of Bits-audited contract that takes no custody of user funds.

Sweden, 14th May 2026 – Poly Syncer, accessible at www.polysyncer.com, has opened general availability of an end-to-end Polymarket Copy Trade engine that replicates the on-chain positions of the most consistently profitable Polymarket wallets to subscriber accounts at a measured p99 end-to-end latency of 1.8 seconds. The engine indexes every active wallet on the venue, re-ranks the field every sixty seconds against a five-factor composite score, and executes orders exclusively through a scoped EIP-712 trading signature that can be revoked on-chain in a single block. The contract has no admin key, no proxy, and no upgrade path; user USDC never leaves the wallet that holds it.

The Structural Gap This Engine Was Built to Close

Polymarket is not a continuous automated market maker. Outcomes resolve discretely to either zero or one dollar, liquidity sits on a central limit order book on Polygon, and the economics of position sizing, hold time, and exit logic diverge sharply from a Uniswap-style swap. Tooling retrofitted from generic AMM copy traders has consistently mis-handled partial fills, CLOB order semantics, and resolution-window edge cases, leaving copied positions stranded or filled at materially worse prices than the wallet they were following. Manual copy trading — a trader watching a leader and entering by hand — typically introduces a thirty- to ninety-second delay, which is long enough to erase the entire edge on thin books and short-duration event markets.

Poly Syncer’s architecture rejects that retrofit. The platform’s published thesis is straightforward: the infrastructure layer for prediction-market execution should be available to anyone who can sign a transaction, the leaderboard underpinning it should be statistically honest, and the platform delivering it should never be in a position to lose a user’s funds. The company has stated in writing that if regulation were ever to compel it to take custody of subscriber assets, it would close before it would comply.

A Three-Stage Pipeline Engineered for Sub-Three-Second Mirroring

The engine operates as three discrete services. A Rust Listener maintains a persistent WebSocket connection to four Polygon RPC endpoints — one premium, three redundant — subscribed directly to the Polymarket CLOB event feed, normalising raw on-chain events into LeaderTrade messages at roughly fifty milliseconds median latency. A Go Risk Engine consumes those messages and applies a layered policy — trust score, liquidity floor, per-trade and per-day risk caps, position sizing, sanity gate — emitting either a MirrorIntent or an explicit rejection at a median of one hundred eighty milliseconds. A Rust Mirror Executor then assembles the matching order under the subscriber’s pre-signed EIP-712 authorisation and submits it through a private Flashbots-style bundle to remove the public-mempool front-running surface. The full pipeline budget — from leader fill to mirrored fill on the subscriber’s wallet — totals approximately 1.6 seconds on the Pro tier and as low as 0.6 seconds on the Elite tier’s co-located node.

The company reports that orders routed through Polygon’s public mempool incurred a measured eleven-basis-point average fill-quality loss against benchmark; the private-bundle routing eliminates that attack surface entirely. Operational integrity is backed by HSM-bound signing keys (FIPS 140-2 Level 3), automated ninety-day key rotation handled atomically through a contract registry, append-only audit logs, and a transparent incident record: one minor event since launch, on 14 March 2026, in which premium-RPC p99 latency rose to 4.2 seconds for eleven minutes following an upstream Polygon validator restart, with automatic failover to the secondary pool, no lost fills, and no funds at risk.

A Five-Factor Composite Score Behind the Polymarket Leaderboard

The platform’s smart money tracker scores every eligible wallet — defined as a wallet with at least ten trades in the trailing thirty days — on a composite weighted as 0.45 · Sharpe-normalised + 0.20 · edge-adjusted win-rate + 0.15 · log-ROI normalised + 0.10 · drawdown resilience + 0.10 · rank stability. The Sharpe component uses log-returns of realised profit and loss against a four-week US T-bill risk-free rate, divided by the ninety-fifth-percentile Sharpe of the daily cohort. The edge-adjusted win-rate subtracts the trade-weighted mean break-even probability — defined as 1 / (1 + odds) — from the wallet’s realised win-rate, so that a sixty-percent win-rate at unfavourable entry prices is correctly recognised as inferior to a fifty-two-percent win-rate at favourable entry prices. ROI uses log-returns clipped at the ninety-ninth percentile and is deliberately under-weighted so no single lucky resolution can vault a wallet into the top of the Polymarket leaderboard. Drawdown resilience combines high-water-mark drawdown, underwater duration, and recovery half-life. Rank stability is computed as the Spearman correlation between a wallet’s daily aggregate rank and its trailing seven-day moving rank, penalising churn that does not reflect repeatable process.

Outlier handling uses a Hampel filter: per-trade PnL observations exceeding 3.5 times the Median Absolute Deviation from the series median are excluded from Sharpe and ROI computations, while preserved in win-rate and drawdown calculations so that actual realised history is retained. The evaluation window is thirty days with a ten-day exponential half-life — short enough to detect a regime shift between election cycles and sports cycles, long enough to avoid sample-size collapse on the typical wallet’s weekly fill count. The methodology is published in full, with explicit formulas, sampling windows, inactivity cutoffs, and an itemised list of known limitations including survivorship bias, regime sensitivity, and partial mitigation of coordinated self-trading clusters.

Variance-Capped Fractional Kelly and Hard Risk Gates

Position sizing is governed by a constrained Kelly formulation: f_mirror = min(α · f*, c_var(σ), c_user), where f* is raw Kelly, α is fixed at 0.25 (quarter-Kelly damping calibrated from a three-year backtest on historical Polymarket fills), c_var(σ) = k / σ² is a per-category variance cap, and c_user is the subscriber-defined maximum percentage of bankroll per trade. The same risk module enforces daily-loss circuit breakers, per-trader allocation ceilings, stop-loss and trailing-stop logic, slippage rejection thresholds, time-of-day and day-of-week windows, and category-level circuit breakers — all server-side, so a subscriber’s discipline is not undone by a phone left at home or an unread alert at three in the morning.

Category-aware filtering covers twenty-five Polymarket verticals — Politics, Sports, Crypto, Finance, Geopolitics, Earnings, Tech, Culture, World, Economy, Climate & Science, Elections, Mentions, Games, Basketball, NBA, Movies, Soccer, Weekly, Recurring, Fed Rates, Business, New Listing, Trending, and Ending Soon — each addressable as an independent gate. A subscriber can mirror a wallet for its Crypto and Fed Rates positions while ignoring its Sports book, directly addressing a common failure mode of whole-wallet mirroring in which a specialist’s casual positions outside their edge erode portfolio performance.

Five Pre-Built Strategies and a No-Code Builder

Subscribers who prefer not to configure their own logic receive five pre-built strategies out of the box. The Top-10 Weighted Basket mirrors the ten highest-Sharpe wallets, weighted inversely by drawdown, with an eight-percent default per-wallet allocation and a twenty-four-hour rebalance cadence. Politics-only Conviction tracks wallets with a trailing ninety-day Sharpe above 2.5 in the Politics and Elections categories and mirrors only trades in which the leader’s position exceeds four percent of their book. Sports Same-Side Fade inverts the signal, mirroring the opposite side of wallets with a negative ninety-day Sharpe in Sports, Basketball, NBA, and Soccer. Earnings Binary Spread activates only during earnings season, splitting each fill across binary legs using Kelly-fraction stakes. Whale-Watcher Copy mirrors any single trade above fifty thousand USDC from top-200 wallets, holding to the source’s exit or seventy-two hours, whichever comes first. Elite subscribers also receive a visual no-code strategy builder with a drag-and-drop canvas — wallet screen, Kelly sizer, category gate, exit rule — backed by a ninety-day backtest preview and parallel A/B variants.

Audit, Bug Bounty, and a Contract With No Admin Key

The execution contract was audited by Trail of Bits in Q1 2026, with scope covering the EIP-712 signature flow, the mirror executor, the risk-engine policy code, and the Safe trading module. The published report records two informational findings — both gas-optimisation suggestions that were implemented before mainnet deployment — and zero medium, high, or critical findings. The on-chain bytecode hash matches the audited artefact. The contract is non-upgradeable, accepts no proxy, and holds no administrator role; revocation of a subscriber’s trading permission propagates within one Polygon block, approximately two seconds, with no operator involvement required.

The company runs a paid bug bounty that pays fifty thousand US dollars for critical disclosures — direct fund theft, unauthorised contract upgrade, signature replay, or key compromise — fifteen thousand for high-severity issues bypassing risk-gate caps, three thousand for medium findings, and five hundred for low-severity defence-in-depth issues. Disclosures are acknowledged within twenty-four hours, triaged within seventy-two, with a ninety-day standard disclosure window and a published PGP fingerprint for encrypted submissions. The company commits in writing not to pursue legal action against good-faith researchers and offers public credit unless anonymity is preferred.

No KYC, No Email, and a Privacy Posture That Begins With Not Collecting

Poly Syncer collects no email address, no name, no government identification, no phone number, no postal address, and runs no Google Analytics, Meta Pixel, third-party advertising SDK, session replay, heatmap, or scroll-depth telemetry. The only persistent identifier associated with an account is the user’s wallet address. IP addresses are cached for sixty minutes for rate-limiting purposes only, are never joined to wallet identifiers, and are deleted at the end of the window. Operational logs covering HTTP traffic, errors, and RPC timing are retained for seven days and then deleted; trade history is retained for the lifetime of an active account and anonymised after ninety days of inactivity. The platform’s stated position is that the cleanest way to honour user privacy is not to collect personal data in the first place.

Wallet screening against the OFAC SDN, EU consolidated, UK HMT, and UN sanctions lists is performed at connection time and refreshed daily; sanctioned wallets are rejected, and active wallets later added to any list are disconnected. Subject-access requests under GDPR and CCPA are verified by wallet signature rather than email or document upload — appropriate for a service that does not hold an email address to verify against.

How Prediction-Market Copy Trading Diverges From Copying a DEX Trader

Copying a DEX trader means mirroring continuous AMM swaps where price discovery is path-dependent and exits are infinite in number. Copying a Polymarket trader means replicating discrete YES/NO event positions with a fixed resolution date and a terminal value of either zero or one dollar. Sizing, hold time, and exit logic therefore behave entirely differently, and an engine that conflates the two will mis-price both. Poly Syncer’s stack is purpose-built for binary-resolution markets — not retrofitted from a generic copy-trading template — and its published latency, slippage, and outlier-handling decisions reflect that specialisation.

Tier Structure and Minimum Capital

The product ships in three tiers, structured so prospective users can validate the leaderboard data before paying. The Free tier offers unrestricted, view-only access to the leaderboard, the smart money tracker, every category dashboard, the read-only strategy templates, and the full methodology. The Pro tier, at $299 per month, unlocks live automated Polymarket trading for up to 250 wallets, unlimited trades, dedicated premium RPC, the complete risk-control suite, Kelly and fixed-fraction sizing, time-of-day windows, private-mempool routing, multi-channel alerting, hardware-key panel security, and CSV/JSON export. The Elite tier, at $499 per month, removes every numerical cap and adds sub-second execution on a co-located node, AI-assisted alpha discovery, an insider-proximity correlation engine, mempool sniping, the no-code strategy builder, an eighteen-month backtesting engine, Flashbots-bundled anti-frontrun routing, hedge mode, cross-market arbitrage routing, governance-vote mirroring, and a raw API and WebSocket signal feed. The minimum capital required to begin live mirroring is twenty-five USDC, with no upper bound, and subscriptions can be cancelled from the dashboard in a single click.

Why a Specialised Execution Layer Now

Prediction markets have transitioned from a niche curiosity into a meaningful pricing venue for political, macroeconomic, and cultural events, with Polymarket volumes repeatedly setting records during major election and earnings cycles. The asymmetry between professional wallets running bespoke trading infrastructure and retail participants navigating the same order book through a browser has scaled in lockstep with that growth. Poly Syncer’s launch is a direct answer to that imbalance: a managed execution layer that any wallet-capable user can switch on, paired with a leaderboard the platform commits to keeping statistically honest, and an architectural posture that ensures the worst case for a subscriber is the cancellation of an EIP-712 permission rather than the loss of custody. By collapsing what was previously a stack of bespoke trading scripts, RPC pool management, and nonce handling into a single non-custodial flow, www.polysyncer.com positions itself as the default infrastructure layer for the next phase of prediction-market participation.

Company Details

Organization: Poly Syncer

Contact Person: Elon Hedlund

Website: https://www.polysyncer.com/

Email: Send Email

Country: Sweden

Release Id: 14052645036

Disclaimer: This announcement is provided for informational purposes only and does not constitute financial, investment, trading, or legal advice. Prediction markets and automated trading tools involve risk, and past performance does not guarantee future results. Users are solely responsible for evaluating suitability, compliance with applicable regulations, and potential financial loss before using any platform or service described.