Multibuy Convergence: How It's Scored, and How It Gets Faked

What Convergence Actually Measures

Multibuy convergence sounds technical. It isn't. It just means several distinct wallets bought the same token inside a short window, and the tooling scores how strong that pattern looks.

A basic score weighs a few things:

• Wallet count — how many separate addresses bought, not how many buys happened.
• Time compression — buys landing inside a tight window score higher than the same buys smeared over an hour.
• Wallet quality — addresses with history, decent PnL, or prior early entries carry more weight than a wallet that's two days old.
• Size relative to the wallet — a wallet going big relative to its own past behavior reads differently than dust buys.
• Whether the wallets have interacted before — shared funding sources, shared timing, shared counterparties.

That last point is where most people stop reading and where the whole signal lives or dies.

The Fake: One Person, Many Wallets

The cheapest way to manufacture convergence is a wallet cluster. One operator funds ten fresh wallets from the same source, waits, then fires buys within seconds of each other.

To a naive scorer that's ten wallets, tight timing, rising volume. It looks like ten people agreed. It's one person deciding.

Red flags that a cluster is staged:

• Funding trace — multiple buyers funded from one address, or from a chain of addresses that all touch a common source.
• Wallet age symmetry — ten wallets created the same day, all dormant until this token.
• Gas behavior — identical priority fees, identical slippage tolerance, identical buy amounts down to the decimal.
• No prior diversity — the wallets bought nothing else, or only bought tokens from the same deployer.
• Sell behavior — clustered entries often cluster on the way out too. Watch whether they exit together.

Convergence without a funding check is just a headcount.

The Fake: Bait Wallets

A second pattern is meaner. An operator uses a few reputation wallets — addresses that have historically been early and profitable — to buy a token they already hold.

The alerts fire. People see "smart money converging" and pile in. The operator sells into that demand.

The tell isn't the buy. It's the pre-existing position. If the wallets buying were already holding before the signal, you're not early to their trade. You're the exit liquidity for it.

This is why wallet quality scores matter less than people think. A good wallet can still be someone else's billboard.

The Fake: Wash Convergence

A third version needs no reputation at all. The operator buys with wallet A, sells to wallet B, buys with B, sells to C — all self-dealing — to inflate unique-buyer counts and volume.

On-chain it reads as activity. In reality no outside money entered. You can usually spot it by checking whether the "buyers" are also the sellers, and whether volume grows while holder count barely moves.

How To Actually Read the Signal

Convergence is a prompt to investigate, not a conclusion. The workflow:

1. Take the token and open it on GMGN — https://gmgn.uk (mirror: https://gmgn.fr).
2. Look at the buyers, not the buy count.
3. Check funding: where did these wallets get their first SOL or ETH?
4. Check history: did they buy together before? Did they sell together before?
5. Check timing against the deployer's activity, not just against each other.

If three wallets bought and you can't explain where their money came from, you don't have a signal. You have a question.

Where Alerts Fit

Convergence alerts are useful because they compress a lot of chain noise into one ping. They are not useful when you treat the ping as a verdict.

Our multibuy feeds exist on each chain — Solana, BSC, and Robinhood — and they're sized for exactly that: a starting point for manual checking. If you want to watch how they fire in practice, the public groups are where the conversation happens:

• CHAT: BH GMGN BASE — @gmgnx_base
• SOL: BH GMGN SOLANA — @gmgnx_solana
• BSC: BH GMGN BSC — @gmgnx_bsc
• ROBINHOOD: BH GMGN ROBINHOOD — @gmgnx_robin

Main alert channels are listed in the directory at https://blackhat.finance/channels.html, and you can load the whole set with the Telegram folder: https://t.me/addlist/1VUQZMhux_JhMzJk.

If you want the scoring vocabulary itself, the metrics reference is at /v2/dyor/reference.html#metrics. The alert conventions are at #alerts, and the baseline rules for reading any signal are at #rules.

The Honest Part

Most memecoins go to zero. Convergence doesn't change that — it just tells you that several wallets bought at roughly the same time. It says nothing about whether the token has a reason to exist, whether the team will dump, or whether liquidity is real.

Use convergence to decide what to investigate. Never use it to decide what to buy. Those are different jobs, and confusing them is how people end up as the exit liquidity for a cluster they never saw.

DYOR. Always.

Community

Stay connected across the chains:

• Blackhat Empire — web terminal, scans and DYOR
• BH GMGN BASE — community, scans, DYOR and shorts
• BH GMGN SOLANA — SOL alert topics
• BH GMGN BSC — BSC alert topics
• BH GMGN ROBINHOOD — ROBINHOOD alert topics
• MAIN alert channels — current public channel directory
• @empiresolanabot — SOL configurable alerts
• @empirebscbot — BSC configurable alerts
• @empirerobinhoodbot — ROBINHOOD configurable alerts

Charts and on-chain research: https://gmgn.uk.