Insider Radar
Fresh wallets taking large, concentrated positions.
The tool runs at /insiders and needs a $49.99/mo membership. There is no free tier and no trial, so opening it signed out lands on the pricing page rather than the board. This page is the open documentation of what it does.
When someone trades on information the market has not absorbed, the trade still settles on a public chain. The behavioural fingerprint is unusually consistent: a wallet created recently, funded, and immediately committing a large position to one side of one market. Insider Radar scans for exactly that overlap.
A flag is a statistical pattern worth a second look. It is never an accusation, it never names a person, and it never asserts that a rule was broken — the tool can see behaviour and cannot see intent.
The three inputs
Wallet age is when the address first appeared on-chain. Under three days scores the maximum, under two weeks scores most of it, under about six weeks scores a little. A brand-new address has no track record to protect and no organic reason to stake its whole footprint on one event.
Concentration is the share of the wallet's open book sitting in its single largest position — eighty per cent or more is the strong band, fifty per cent the weaker one. Size is the absolute stake, with bands at ten, twenty-five and a hundred thousand dollars. Any one of these is common on its own; the rare overlap of all three is the signal.
How the score is built
The three components are weighted — age most heavily, then concentration, then size — and mapped onto a one-to-ninety-nine scale. A candidate is only listed when it clears forty and at least one named reason actually fired, and the reasons are printed on the row so you can see which of the three did the work.
The minimum-signal control on the page is a filter over the results, not a rerun of the scan, so lowering it can only ever widen the list.
Why the scan runs at a fixed low floor
The scan reads the exchange's large-trade tape at a fixed low dollar floor, de-duplicates to one entry per wallet and market, keeps the largest candidates, and then enriches each with profile age and position book. Your dollar filter is applied afterwards.
That ordering exists because of a real bug: when the dollar floor was pushed into the tape query itself, asking for a lower floor returned a tape of small very recent trades whose row limit no longer reached back to the older large prints — so lowering the floor could shrink the list. One canonical scan plus a pure filter makes the results monotonic, which is the behaviour anyone would assume.
What to do with a flag
Read the position in full — which market, which side, how large, how fresh the wallet — then ask why it would exist. The pattern carries far more meaning in an obscure, information-sensitive market where a small number of people would plausibly know something first than in a heavily covered election where thousands of eyes are already on the price.
Then cross-check before committing anything. Watch the address on the live feed, read its whole book, and set an alert rule if you want the next unusual move from it to reach you. A flag is a question, and the traders who get value from it treat it that way.
What it cannot tell you
Every tool has a boundary, and knowing where it sits is the difference between using one well and being misled by it. For Insider Radar:
- It reports a statistical pattern, never an accusation, and never a person. A wallet address is not an identity.
- Fresh wallets have innocent explanations: privacy-conscious traders rotating addresses, genuine first-time users, market makers, and funds moving between wallets all trip the same wires.
- Polymarket only. Kalshi has no public wallets, so address-level detection is impossible there by construction.
- Concentration is measured on the open book at scan time. A wallet that has already exited most of its other positions will look concentrated for a boring reason.
- The scan covers a bounded candidate window of the largest recent prints, not every trade on the exchange. Absence from the board is not evidence of absence.
- The chain shows the position and never the reason for it. Every interpretation downstream of a flag is yours, and uncertain.
What it is built on
- Polymarket data-api large-trade tape, filtered by cash size, as the candidate pool.
- Polymarket public profile data for wallet creation time.
- Polymarket data-api /positions for each candidate, which is where concentration is computed.
Related reading
- Polymarket Insider Radar: How to Track Fresh-Wallet Whales Before the Market Moves — Guide
- How to Spot Insider Trades Before the News Breaks — Strategy
Tools that pair with it
Large trades from scored wallets, streaming across both venues.
Just-listed markets, before the flow finds them.
One market, every signal we hold on it, on one screen.
WhaleTracks is informational analytics, not financial advice. Market data comes from Polymarket, Kalshi and Manifold's public APIs; WhaleTracks is not affiliated with any of them. Past performance does not guarantee future results.