Field manual for the trenches

Anyone can create a token in about forty seconds. No code, no money, no permission. That is the whole point — and the whole problem.

There is no order book at the start. A formula decides the price: each purchase pushes it up, each sale pushes it down. When roughly $69,000 of buying accumulates, the token graduates — it moves to a real exchange and the curve stops mattering.

You do not need to launch anything. But you need to know exactly what a creator can do, because every trick below will be used on you.

Step five is where honesty is decided. Dev buy means the creator buys his own token before anyone else sees it. Small amount: normal, he wants skin in the game. Large amount: he owns the supply and you are his exit.

GMGN shows what the chart hides. You are looking for four numbers. One bad number ends the conversation — you move on to the next token.

Four out of five numbers are bad. The chart for this token looked perfect at the time — vertical green candle, hundreds of buys. That is exactly what a bundle looks like from the outside.

Videos show someone hitting 100x every other day. A real session looks like this, and it is worth knowing before you start — otherwise you will think you are doing it wrong.

The hard part is not finding a token. The hard part is saying no fifty-five times in a row when any one of them could have been the one.

One wallet for everything is how people lose years of gains in a single click. You need three, and they never talk to each other.

A rug pull is not a mystery. It follows a script, and the script has tells you can spot before the red candle.

Every one of those numbers was visible on GMGN before the pull. The chart said buy. The data said run. The data was right.

Finding a good token is maybe a third of the job. The rest is leaving — and almost nobody plans it before entering.

After selling half at 2×, your original money is back in your pocket. Everything after that is playing with winnings — and that changes how you think. You stop panicking at every red candle.

Most people do not lose because they picked badly. They lose because one bad pick was too big.

That is the whole argument. Losses do not scale linearly — a big one is mathematically much harder to undo than it feels.

A meta is whatever theme the market is buying this week — cats, politics, AI agents, a viral clip. It always follows the same arc.

By the time a meta is obvious enough to read about, you are late. The copies almost never work. If you missed the first one, the honest move is to skip the theme entirely.

Wallets are rarely broken into. People are talked into opening them. Here is what that looks like in practice.

A wallet is not an account. Nobody can reset it, nobody can recover it, there is no support line. It is a key — and whoever holds the key owns whatever it opens.

What actually drains a wallet

Every one of these takes roughly 1% per trade. The differences are speed, what they show you, and whether you actually hold your keys. Fees are almost identical — pick on workflow, not on cost.