Thirty-seven seconds. That is how long the gift took to arrive. The tokens were bought by one wallet. All 650 million of them landed in the wallet of the trader they were named after. Our June snapshot caught the split. This piece is what came next: the pump, the drops, the bleed, and who ended up holding the bag.
We followed the token for its first four months through Bitquery's indexed Solana history. Every trade, every failed swap, every token move, wallet by wallet, down to vault sweeps that price feeds never see. The ledger below is the result.
01 · The minute that made itA $6,275 buy, a gift, an exit
The whole opening sequence lives on-chain, and it takes less than a minute to read. The launch trade pays 84.74 SOL for 792.45 million ANSEM, 79.2% of the supply, through pump's launch sale. The same second, a PumpSwap pool is born, seeded with a fifth of the supply and 85 SOL. Then comes the gift, then the exit: over the next nine seconds the buyer sells the rest of its tokens into its own pool and is done, clearing about twelve thousand dollars.
Two details make the minute interesting. The wallet that signed the buy is no fresh account; it has been trading since April, with a long history across 2,806 tokens. Whoever ran this launch had run things before. And the wallet that received the gift is a genuine working account, busier than most, that has traded hundreds of tokens since. It never bought ANSEM in size, and it has never sold one.
02 · LimboEleven quiet days
Then, for eleven days, nothing. Volume ran between one and nine hundred thousand dollars a day, price drifted around $0.0001, and the only people watching were a few hundred accounts a day, plus one June 19 crowd of nearly 18,000. A token named after Solana's loudest bull sat in the trench it was born in. Our own June piece read this window as a dead coin waiting for its namesake. The full tape shows the pool was live and open to all from the first minute, with real volume and hundreds of accounts from day one. It was simply quiet.
03 · The embraceJune 27: the story changes
On June 27 the namesake publicly claimed the token and attacked pump.fun's creator-reward structure, promising to hand back what he could: a stimmy for the trenches. Volume exploded a hundred times in two days, and daily traders went from a few hundred to tens of thousands.
The handouts came in two shapes. The next tranche after the claim went to nine wallets a full day before anything reached the public, and those wallets were not the trenches. One sold its entire eight-million-token allocation 44 seconds after receiving it. By late August the group had fed most of its tokens back into the pools for cash. Only the following day's drop was a real gift, spread across hundreds of wallets, with a chain of vault-style drop bots paying small claims to 75,000 more wallets through August.
One of the nine later became the wallet Lookonchain publicly flagged: it sold for a fraction of what the stack was worth three weeks later at the top. Early money has its own logic, and it is rarely patience.
04 · The arcThree weeks up, three months down
Price peaked in the first hours of July, a 3,200× return from launch in three weeks, with the market cap grazing a third of a billion dollars on a one-billion supply. Reporting at the time put his whole bag up nearly two hundred million on the week, and his wallets agree: the position alone gained over a hundred million that week without him touching it.
Then the long slide. August held for a while around $0.25, bounced once, and broke. By early October the price had given back seventy percent of the top and the daily crowd was down to 10% of its peak size. The token still trades real volume every day, which is its own small miracle.
05 · The ledgerWho actually made money
We rebuilt every wallet's position from the raw transfer ledger, marking unsold tokens at the October price. The crowd nets out slightly positive. The head count is brutal: three in five wallets lost money. And the profits pool the way they always do in these markets; a hundred wallets hold the entire net gain between them, and everyone below breaks even.
The crowd itself was a tourist crowd. 23% of all wallets made exactly one swap and never came back. More than 120,000 ran the full round trip, buy to final sell, inside an hour. Only one wallet in nine stayed active past a month. And the tape underneath was mostly machine-made: half a thousand high-frequency wallets moved a third of all volume, and on the peak day the failed-swap log recorded nearly three failed attempts for every trade that landed, the classic mark of MEV bots fighting over the fresh pools.
One blind spot survives any swap-based audit: the LPs, the people who funded the pools. Three of the ten biggest winning accounts never bought a token on any DEX. The largest received 10% of the supply straight out of the main Meteora pool, one huge LP pull, then sold part of it into the market. Swaps do not see the LP game, and the quietest big winners played it. The main pool had fewer than 2,000 LP signers, and one program alone handled most of its LP moves, a single machine doing the work a crowd is assumed to do.
06 · The believersWhales still on the board
August brought a different kind of buyer. Four wallets put in between three and a half and seven and a half million dollars each, around a quarter to a third of a dollar average, and none of them has sold. The largest swept its whole position into a fresh cold wallet and has not touched it since. Together they hold tens of millions of tokens bought well above today's price, marked down roughly two-thirds. These are not tourists. Somebody with size decided this token was worth holding through a 70% drawdown, and so far they have.
The peak also left its own residue: of the wallets that first bought during the three days around the top, 40% still hold, and a few hundred of them hold stacks bought within cents of the high.
07 · The 58% questionOne wallet is the market
Strip everything else away and one fact decides this token's future: the namesake's wallets still hold 58% of the supply, split between an active wallet and a cold wallet that has not moved since August. He has never sold. He has also never locked it, with no vesting contract, no timelock and no burn. RugCheck flagged the stack as a price risk in July, and the older context is a 2024 claim that he promoted low-cap tokens to his audience, unproven and denied.
08 · The lotteryThe copycat economy
The name spawned its own economy: 48,565 sister tokens carrying some version of the name, which together did more swaps than the original and lived, almost all of them, for days. The survival lottery is harsher than any fraud stat: of the 389,895 tokens launched on pump.fun in the same two weeks as $ANSEM, 91 were still trading a hundred days later, and this token is the only one of its cohort to pass a million swaps. Every vanishing twin of this launch was somebody's bag.
09 · Close of ledgerWhat four months produced
One anonymous spend of $6,275 produced a peak market cap near half a billion dollars, a billion and three quarters in trading, half a million trading wallets, 48,565 copycat launches, and a trading crowd that nets positive only because the early cohorts and the drop economy cashed out before the music slowed. The one player who never put a dollar in, the man whose name is on the token, is the only one still sitting on a nine-figure stake.
The ledger does not moralize. It just shows, wallet by wallet, who paid, who collected, and who is still holding. Run the queries yourself, or read our snipers-and-bundles piece to see the same mechanics at the launchpad level.
Audit any token yourself
The indexed Solana data behind this four-month audit, from full transfer ledgers to failed swaps and per-wallet P&L, is queryable the moment a block is final, across 40+ chains with one API key.