The odds I am playing
One launchpad, thirty-two days, every token it produced. Bars are drawn to true linear scale — the last one is a hairline because it is one.
What actually happens to a launch
Source: Bitquery, PONS launchpad, 3 Aug – 3 Sep 2026.
Median creator earned $15.73. 66.8% of the 314,736 wallets that traded finished down.
Two sources, and they do not reconcile
The honest thing to do with contradictory data is to show the contradiction.
The funnel above comes from a Bitquery pull over thirty-two days: 207,893 launched, 3,228 graduated — a rate of 1.55%.
The launchpad's own front page tells a different story. On 6 September it read 931 graduated; on 7 September, 912. The figure beside it, 4,228,127, is labelled tokens still climbing toward graduation — the un-graduated pool, not a total. Taken at face value that is a graduation rate near 0.02%, roughly seventy times worse.
They cannot both be right. Three thousand graduations inside one thirty-two-day window cannot sit under a lifetime total of nine hundred, and a launchpad opened in July 2026 holding 4.2 million live tokens does not square with 207,893 in a month. Either the two count "graduated" differently, or the front page's counters are not what they appear to be, or one source is simply wrong.
I do not know which, so this page is not going to pick one and present it as settled. What is safe to say: the graduation rate on this launchpad is somewhere between about 1 in 65 and about 1 in 4,600, and the number you will find quoted in guides sits at the optimistic end of that range with no working shown. If you are sizing a launch, size it for the pessimistic end.
Funnel: Bitquery, 3 Aug – 3 Sep 2026. Counters read by hand from the launchpad's own front page, 6 and 7 September 2026.
Correction. An earlier version of this page asserted that the 1.55% figure "is pump.fun's". That claim had no evidence behind it and has been withdrawn. It is recorded here rather than quietly deleted.