AI

Multibuy Convergence: How the Score Is Built, and How It Gets Faked

Multibuy convergence is a timing signal, not a verdict. Here is how the score is assembled and the cheap ways it gets gamed.

· 7 min read · Blackhat Empire

What the score is actually measuring

Multibuy convergence is shorthand for one idea: separate wallets start buying the same token inside a short window without a visible coordinating thread. The score is a compression of that idea into a number. It is useful. It is not a verdict, and it is not close to impossible to fake.

If you treat convergence as "smart money arrived," you will get farmed. If you treat it as "something is pulling attention here, now go verify it," you stay alive. Most memecoins go to zero. The ones that don't still eat people who confuse a signal with a guarantee.

The ingredients that usually go into it

Different tools weight these differently, but a convergence score generally blends a few measurable pieces:

  • Wallet count inside the window. Three buyers is a coincidence. Twelve is a pattern.
  • Wallet quality. Fresh wallets with one trade versus wallets with history, realized profit, and no shared funding.
  • Timing density. Buys clustered in seconds versus spread across an hour. Tight clustering scores higher and is also easier to manufacture.
  • Size relative to liquidity. Large buys into thin pools move price more and score more.
  • Independence penalties. Shared funding sources, identical gas patterns, or wallet creation batching should reduce the score. Whether a given tool actually applies them is the question you need to ask.

Notice what is missing: nothing about why they are buying, and nothing about whether the token survives the week. See the metric definitions at /v2/dyor/reference.html#metrics for the underlying terms.

Where the faking happens

Wallet farming. One operator splits capital across dozens of wallets, funds them from the same source at the same time, and buys in sequence. To a naive counter, that is twelve independent buyers. To anyone checking funding trees, it is one hand.

Wash-style pre-loading. Buys are placed early and small to seed the count, then a larger buy lands once the score is live and visible to people watching. The convergence was manufactured to attract the trade that follows.

Timing games. Buys fire in the same block or the same second. Real independent humans rarely move with that precision. Tight clusters should be a warning as much as a signal.

Score-chasing bots. Some groups run strategies designed purely to trip convergence thresholds on tools they know traders watch. If a score can be gamed by anyone who knows the formula, assume someone is doing it.

Fresh-wallet noise. New wallets are not automatically suspicious — every real buyer was once fresh. But a cluster of wallets created within the same hour, all buying the same token, is not organic discovery.

How to pressure-test a convergence alert

When a convergence signal lands, do not buy the alert. Do this instead:

  1. Open the token on GMGN at gmgn.uk (mirror: gmgn.fr) and look at holder distribution and buy timing.
  2. Check funding. Do the buying wallets share a source, a creation window, or identical gas behavior?
  3. Check liquidity depth. A high score into a shallow pool is a story about thin liquidity, not conviction.
  4. Check what happens after. Do the buyers hold, or does supply start landing on the ask within minutes?
  5. Check the token's reason to exist. If the only narrative is the score, the score is the product being sold to you.

If you cannot answer those in under two minutes, the honest move is to pass.

Using the score without worshipping it

Convergence is best read as a prioritization tool. It tells you where to look first, not what to buy. Treat a high score as a reason to open the chart and the holders, and treat a low score as a reason to keep scrolling.

It also helps to see who is flagging what. Our MAIN alert channels are organized by chain, so you can watch how a signal develops in context rather than as a single number: SOL alerts at @BlackhatAlerts, BSC at @empirebsc4volume, ROBINHOOD at @empirerhboosts, BASE at @gmgn_base. Chat lives at @gmgnx_chat. The full directory is at blackhat.finance/channels.html.

None of that replaces your own read. Alerts are inputs. Your eyes are the filter.

The rule that survives every score

Any number that can be computed can be gamed. The defense is not a better number, it is a verification habit: funding trees, timing, liquidity, post-buy behavior, and a clear exit plan before you enter. Our house rules at /v2/dyor/reference.html#rules exist for exactly this reason, and the alert conventions are explained at /v2/dyor/reference.html#alerts.

Multibuy convergence is a decent map of attention. It is not a promise, not a floor, and not a reason to size up. Most of these tokens die. The ones that don't still punish people who skip the checks.

Community

Stay connected across the chains:

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