On pump.fun, your first launch is your best launch
A deployer's first coin graduates 2.67% of the time. Its 101st graduates 0.85% of the time. The gap between launches, the launch count and the last graduate all say the same thing.
Quick answer
A pump.fun deployer's first launch graduated 2.67% of the time. Its second to fifth graduated 0.97% of the time, and everything from the 101st launch on graduated 0.85% of the time. The 5,647,128 launches in the tables that were somebody's 101st or later (37% of all launches) came from wallets that had already graduated a coin 72.1% of the time, and it did not help them: the deployers who learn the most launch the worst coins, by the only test the curve applies.
The nth launch
Every launch in the tables knows how many launches its deployer had made before it. Sort by that.
| the deployer's launch number | launches | graduated | graduation rate | the deployer already had a graduate |
|---|---|---|---|---|
| first | 3,898,141 | 103,911 | 2.67% | 0.0% |
| 2nd to 5th | 2,041,624 | 19,753 | 0.97% | 3.6% |
| 6th to 20th | 1,738,993 | 15,116 | 0.87% | 8.3% |
| 21st to 100th | 2,066,141 | 17,964 | 0.87% | 20.9% |
| 101st or later | 5,647,128 | 47,862 | 0.85% | 72.1% |
First launches are a quarter of all launches and half of all graduations. The rate falls off a cliff after the first and then goes flat: from the sixth launch to the ten-thousandth, a coin graduates about one time in 117, whoever the deployer is.
Two things are mixed in that first-launch number. One is the wallet that launches once because it is a project: a team, a community, a person with an idea, who put real money behind one coin and never launched another. The other is a fresh wallet used once by an operation that launches from a new wallet each time, which is the opposite of a project. The table cannot tell them apart; the deployers report has the ones who launched tens of thousands of coins from a single wallet, which is the other extreme.
The gap between launches
Launches by a deployer who had launched before, sorted by how long it had been since their previous launch.
| time since the deployer's previous launch | launches | graduated | graduation rate |
|---|---|---|---|
| under a minute | 2,837,884 | 9,569 | 0.34% |
| 1 to 10 minutes | 4,184,794 | 37,062 | 0.89% |
| 10 to 60 minutes | 1,994,877 | 20,394 | 1.02% |
| 1 to 24 hours | 1,512,903 | 19,068 | 1.26% |
| 1 to 7 days | 581,860 | 8,912 | 1.53% |
| 7 days or more | 381,568 | 5,690 | 1.49% |
A quarter of all repeat launches came within a minute of the same wallet's previous one, and those graduate at 0.34%: a third of the rate of a launch that waited an hour, a fifth of one that waited a week. A coin launched forty seconds after the last coin from the same wallet is not a coin anyone meant.
Among the heavy deployers alone (100 or more launches before this one), the ordering is the same and the levels are a little higher: 0.40% under a minute, 1.81% after a day or more.
Having graduated before helps, a little
Repeat launches, sorted by how many of the deployer's earlier coins had graduated.
| the deployer's graduates before this launch | launches | graduated | graduation rate |
|---|---|---|---|
| none | 6,775,220 | 26,599 | 0.39% |
| one | 1,035,558 | 13,833 | 1.34% |
| 2 to 4 | 1,026,229 | 15,704 | 1.53% |
| 5 or more | 2,656,879 | 44,559 | 1.68% |
A deployer with five or more graduates behind it graduates its next coin 1.68% of the time, four times the rate of a repeat deployer with none. That is the whole edge of a track record, a few percentage points at best, and still below the 2.67% of a wallet launching for the first time.
The wallets that launch in batches
The fastest heavy deployers launch in batches. Among wallets with 100 or more launches, the shortest average gaps between launches are zero and one second: AbbAbjTQxszFWG82tMwxqWppaX7UUtf7GCRNU1JEveRL launched 1,481 coins at an average of one second apart and graduated one of them; 7Gjki2Kin1XoCrdJCc1kBQn914RLwZGviKs9pNpdpQMS, 303 coins, one second apart, none. These are scripts that create coins in the same transaction or the same block. Across all deployers with 100 or more launches, 1,706,670 launches came within a minute of the previous one and 6,812 of those graduated.
What we counted
- The tables as of 2026-10-16, built from our archive of every pump.fun trade: 15,392,027 launches from 2025-03-07 to 2026-10-16 by 3,898,141 deployer wallets; 204,606 graduations.
- A deployer is the wallet that signed the create. Launches before, graduates before and the gap are counted within the tables, from each launch's own history at the time it launched, by block time.
- Graduated: migrated to PumpSwap at any later time.
- "Already had a graduate": the share of launches in the row whose deployer had at least one graduated coin before that launch.
- The gap tables cover launches with a previous launch by the same wallet (11,493,886 of them); first launches have no gap.
- The batch wallets: average gap in seconds between consecutive launches, over wallets with 100 or more launches with a gap.
Frequently asked questions
Does a deployer's track record predict a graduation?
A little. Repeat launches by a wallet with five or more earlier graduates graduated 1.68% of the time; by a wallet with none, 0.39%. A first launch from any wallet graduated 2.67% of the time.
How fast do serial deployers launch?
A quarter of all repeat launches (2,837,884) came less than a minute after the same wallet's previous launch. Those graduated 0.34% of the time.
Is a first launch a good sign?
It graduates at the highest rate in the table, 2.67%, but the row mixes one-coin projects with fresh wallets used once by operations that never reuse a wallet. The deployer's record is one input, the bundle and the five-minute market cap are others; the bundles and first-24-hours reports cover those.
How do I check a deployer's record?
GET https://pumpdata.fun/launch/{mint} returns the deployer's launches, graduations, best run and the gaps between launches, with the bundle at creation and how launches that looked the same at this age went over their next 24 hours. $0.005 a call.
Every number here comes straight from the blockchain, the same data behind the paid calls: the calls ($0.005 to $0.01 each, USDC over x402, no account), yesterday's /stats and the daily wire are free.