Methodology — how this works

Smart-money consensus for prediction markets. Intelligence, not auto-betting.

What this is

We surface where the most profitable Polymarket traders independently agree. The thesis: the edge is consensus among proven winners, not chasing any single whale. A lone whale's trade is noisy and you're always too late to copy it cleanly — but when many of the top-ranked, consistently profitable traders back the same outcome, that agreement is a slower-moving, higher-conviction signal you can actually use.

How traders are chosen

• The top ~300 by monthly profit on Polymarket's leaderboard, per market category — the ranking itself is the quality bar (the #300 trader still cleared tens of thousands in profit).
• Ranked by profit, not volume. A high-volume churner is noise; a consistently profitable account is signal.

How to read the grid

Rows are ranked by how many top traders bet them, so the biggest-conviction games rise to the top. Sports group by discipline (Tennis, Soccer, Baseball…); Politics and Crypto show as a single section each.

How we're different

There's a real, crowded field of Polymarket "whale trackers" now — some good, some paywalled, some spam. We're honest about it: showing trader consensus is no longer a unique idea. Here's what actually sets this apart, and what it doesn't do.

What we deliberately don't do (yet): we're an analysis and intelligence board, not a sub-second execution feed. We don't push real-time Telegram/Discord alerts the instant a whale moves, and we don't auto-bet. If you need millisecond alerting, a paid alert product fits that better — we're built for higher-conviction, slower-moving consensus you can reason about.

Our stance — why we're not the spam

There's a crowd of "copy the top traders" products. They range from real-but-losing to outright wallet-drainers. We are deliberately different:

• Intelligence first. We show where smart money agrees and let you decide. We are not a one-click auto-copy bot — auto-copying a single whale structurally loses to slippage (by the time you see the trade, their order already moved the price).
• Non-custodial, always. We never take your funds, never ask you to deposit, never ask you to deploy code, never request wallet approvals.
• No promises. We never claim "risk-free", guaranteed returns, or "X% weekly". Consensus is a conviction signal, not a guarantee.
• Transparent. Every trader is a public Polymarket wallet; every bet is on-chain verifiable.

Limitations (the honest part)

• Reflects recent activity (a rolling window per category) — it's a snapshot of what smart money is doing now, not a tip sheet.
• Sparse on quiet days. If the top traders aren't active in a market, the grid is thin — that's honest, not a bug.
• Markets we can't classify are excluded. Past performance never predicts future results.
• This is information, not financial advice. Bet responsibly.

Corrections (we publish our own bugs)

2026-07-14 — consensus miscount on multi-market events, fixed. Some events (e.g. a soccer match) are several separate Yes/No markets under one event. A netting bug merged Yes-buys on opposite teams into a single fake "consensus" figure. The board logic was corrected, and the 86 paper picks recorded from those shapes before the fix are now excluded from the headline record and the edge scanner — shown separately, and preserved unchanged in the tamper-evident ledger. The same day we rebuilt the resolve hash chain after a stamping-order bug made link verification report a false failure (no pick data changed; both chains verify green).

See our live paper-trade record → — every high-consensus play recorded + scored when it resolves, wins and losses both.

Data: public Polymarket Data API. No login, no custody, no execution.