← blog · 2026-09-09 · 3 min read
Synced buys, empty pools and copycat tickers: the launch patterns that catch copy-traders
Most bad fomo.family launches are not honeypots. They are coordinated buys designed to trigger copy-trading, pools with no real money behind the headline liquidity, and lookalike tickers. Here is how each works and how hoodwatch flags it.
A honeypot check answers one question: can the token be sold? Most losses on fomo.family come from tokens that pass that check. The pool is sellable; there is just nobody to sell to, because the buying you copied was staged. These are the three patterns we flag, with the exact rules.
Synced buys
A launch operator controls several wallets that are followed on fomo. Within a couple of minutes of the pool opening, three or more of them buy the new token in near-identical sizes. Followers see three respected traders "converging" and pile in. The operator sells into that wave.
The tell is precision. Real traders buying independently do not land within 90 seconds of each other with sizes that differ by a few dollars. hoodwatch's rule:
- three or more tracked wallets make their first buy of the token within three minutes,
- and the sizes have a coefficient of variation below 0.15.
The synced buys flag becomes final 15 minutes after the first buy. On the DEAL launch that prompted the rule, three tracked wallets bought $177 each inside two minutes, and the pool had $487 of ETH behind a five-figure headline liquidity.
Empty pool
Aggregators report liquidity as both sides of the pool. On a fresh launch the token side is priced at the current tick, which the operator sets. A pool can show $20k of "liquidity" and $50k of market cap with under $2,000 of actual ETH in it.
hoodwatch computes real liquidity as reported liquidity minus the token side (base amount times price) and flags an empty pool when real liquidity is under $2k behind a headline over $20k or a market cap over $50k. The paper trading desk refuses entries below $5k real liquidity, and the triage board requires $10k for a ticket.
Copycat tickers
When a token runs, lookalikes launch within the hour: the same ticker, the ticker plus a letter, the ticker with one character changed. Copy-traders searching by name buy the wrong one. hoodwatch compares every new token's symbol against tokens that had six or more buyers in the last 24 hours and flags a copycat when the symbol matches exactly, contains the original, or is one edit away, and the new token launched after the original started running.
A copycat is not always a scam; sometimes it is a derivative play in its own right, and we show a separate derivative window when the original is still running. But a copycat of a runner that has stopped running is exit liquidity for whoever launched it.
Deployer memory
None of these patterns exist in isolation. The same deployer or owner address that launched a honeypot last week launches the "organic" token this week. hoodwatch keeps a history per deployer and per owner: two to fifteen tokens with a honeypot or three or more rugs of 25 percent or worse earns a bad deployer flag on every new launch from that address.
How the flags are used
On the tape each flag is a tag on the row. On the triage board a token with any of these flags goes to the SKIP lane regardless of who is buying. In research reports they deduct from the tape score before the language model ever reads the website. The point is that the model should never be able to talk itself into a launch the chain already says is staged.
Next: What 30 days of fomo tape data say about following the crowd.
hoodwatch is unofficial and read-only, not affiliated with fomo.family or Robinhood. Nothing here is financial advice. Open the live tape.