How insiders exit, and what it looks like on the tape while it happens
Short answer Insiders rarely exit with a crash. They sell into demand while the price holds. Learn the tape pattern, what the Scanner reads, and the one flag that matters.
You want to know whether the people who got in first are getting out. A chart cannot tell you their intent. A wallet selling is not proof that it belongs to the team, and a wallet holding is not proof that it is committed. You can read two other things: whether the first buyers are able to leave, and whether the flow on the tape shows them doing it. Those two questions have answers on any chain.
Launch bundling, in plain terms
The cheapest supply in a token's life is bought in the first seconds of trading. Bundling means one operator using many wallets to buy inside that window, sometimes all in one block, sometimes across the first handful of blocks. The point is to take the earliest price with the largest size, spread across addresses that do not look connected.
A bundle is a fact about supply, not a verdict on the token. It tells you that a group holds a large slice at a cost far below the current price. The next question is what that group can do with it.
Three states for the early supply
Early supply sits in one of three states, and each one reads differently on the tape.
Still held, and able to leave. The wallets are unlocked and the pool is deep enough to pay them. Each of those wallets is a seller waiting for a price. The price on your screen has not met them yet.
Still held, and unable to leave at this price. The group's bag is worth more than the pool could pay out. This is the state that gets misread. Some people call it "trapped supply" and treat it as calm, on the theory that supply which cannot sell does not press on the price. The Scanner flags it instead. A bag the pool cannot pay out does not sit still. It gets sold into every rally, in pieces, for as long as buyers keep arriving, and the pool thins with each sale. The Scanner's exit question answers "Not at anything like the price on the screen" when it finds this, and marks it as danger.
Already sold. The early wallets have cleared most of their position. The overhang is gone. Whoever holds now paid closer to the current price, and the question moves on to whether anyone wants the token at all.
The useful distinction is not how much insiders own. It is how much of what they own can reach the market, and what they have already done with it.
The tape pattern that shows distribution
Distribution means early holders selling into demand. It rarely looks like a crash. It looks like absorption.
Price holding or drifting up while sells keep hitting. Somebody is buying everything offered, and the seller is content to keep supplying it.
Repeated similar-sized sells. A wallet selling a fixed slice at intervals is running a schedule.
Top-holder balances shrinking while the price stays flat. The holder count can look stable while the share held by the earliest wallets falls.
New wallets arriving as the earliest wallets send tokens out. Fresh buyers meeting old supply.
Volume rising while price does not. Often the sound of transfer rather than accumulation.
Each of these looks healthy on its own. Rising volume with a stable price reads as consolidation on a chart. Only the composition of the flow separates the two, and that means looking at who is selling, not how much.
What the outcome looks like, and how the Scanner reads it
The Scanner's full read asks "Has the smart money already left?" and answers it from the bundle scan and the behaviour of the coordinated wallets since launch. It gives one of a few answers.
If the wallets that bought together at launch have already sold, the read says so. The overhang is gone and the answer is about who holds now.
If they are selling right now, the read says "They are selling right now, into whoever is buying." That is distribution in progress, and the read marks it as danger.
If they are still holding, the read says so and, where it could price the pool, tells you how far the token would need to run before selling would pay them back. A group that needs several times the current price before a sale covers its cost has a reason to wait. A group already in profit does not.
If the launch could not be read, the answer is unknown, and the page says that an unreadable launch is not the same as a clean one.
Under those answers, the bundle read lists who funded the early buyers, how much supply those wallets still hold, whether they have sold, and what the pool could pay them at today's price. The holder read counts what the top ten wallets hold with pools left out, so that number will be lower than a raw holder list elsewhere.
Our logs give a sense of how much the bundle read mattered at entry. Among Solana calls where the first bundle scan found a confirmed bundle, 14.4% went on to reach double the market cap where we logged them and 62.5% went flat or down (n 2,567). Where the scan found no bundle, 24.3% doubled and 49.7% went flat or down (n 4,102). Those are as recorded in our logs, with scans taken a median of under a minute after the sighting, and the live figures sit on /research.
When the crowd arrives
Early holders need buyers. The moment a crowd shows up is the moment a schedule of sells has someone to meet. The Trade Bot's FOMO alerts, which you turn on under Settings, send one message when at least your chosen number of FOMO traders buy the same token within an hour, under a market-cap ceiling you pick. Read that alert two ways. It tells you attention has arrived. It also tells you that anyone holding cheap supply now has a bid to sell into. If you hold, that is a time to check the smart-money read again rather than to add.
What this does not tell you
The read cannot name the people behind the wallets. "Coordinated" means funded or timed together, not identified.
A clean smart-money read describes the launch wallets. A later buyer with a large position can distribute in the same way, and the bundle read will not show it.
The read is a snapshot with a time stamp. Wallets that were holding at 14:05 can be selling at 14:35.
Nothing here predicts price. A token whose insiders have sold can still fall, and a token mid-distribution can still rise while the buying lasts.
What to do next
Paste the contract at /scan and open the full read. Read "Has the smart money already left?" before you read the chart, and if the exit question is marked danger, treat the market cap on the screen as a number the pool cannot pay.
What the video says the full transcript, for reading or searching
The lesson is the method. The product is where you apply it.
Paste any Solana or Robinhood Chain contract into the PumpPill trade bot: you get the risk read first, then buy or sell in one tap, with limit orders and auto-sells.