Only one in 58 tokens launched on Pump.fun reached a full trading pool over the past 11 months, while 79% never traded again after their launch day, according to a new Bitquery investigation that tracked 11.5 million new Solana tokens.
The research, published October 10 using on-chain figures verified through October 7, provides one of the clearest measurements yet of what happens after the constant stream of new tokens appearing across Solana launchpads.
Bitquery found that roughly 11.5 million new tokens traded on Solana between November 1, 2025 and October 7, 2026, equivalent to about 32,000 on the median day. Pump.fun accounted for 78% of those tokens, making its launch data particularly significant for understanding the economics of the broader Solana memecoin market.
The survival numbers are severe. Half of Pump.fun launches recorded their final trade within roughly three minutes of their first trade, while about one-third were finished within a minute. Overall, 79% never traded beyond their launch day.
Longer-term survival was rarer still. Only 1.5% of Pump.fun tokens recorded any trade during their fifth week, while just one in 437 managed at least 10 trades during that period.
The findings come from Bitquery’s investigation of Solana launchpads, which reconstructed the lifecycle of tokens from their first decentralized-exchange trade through their subsequent activity or final trade.
Only 1.7% of Pump.fun Launches Reached a Full Trading Pool
Pump.fun begins new coins on a bonding curve, where buying pushes the price upward and selling moves it lower. Once sufficient capital enters the curve, a token can graduate to PumpSwap, where the accumulated SOL and remaining tokens are migrated into a conventional liquidity pool.
Bitquery found that only one in 58 Pump.fun launches completed that process during the study period, equivalent to roughly 1.7%.
The graduation rate did improve substantially during 2026. Tokens launched in November 2025 graduated within 30 days at a rate of about one in 144. For August launches, the figure improved to roughly one in 32.
But graduation itself has also become easier to manufacture quickly. In November, only 2% of graduations occurred after three bonding-curve trades or fewer. By September, one-third of Pump.fun graduations had reached the threshold in three trades or fewer.
That weakens the usefulness of graduation as a standalone signal that a token has attracted a large organic trading community.
It does not make graduation meaningless. Bitquery found that 43% of graduated Pump.fun tokens were still trading during their fifth week, compared with only about 1% of those that never graduated. A completed graduation therefore still correlates strongly with survival, even though getting there does not necessarily demonstrate broad demand.
A Small Group of Wallets Produced Almost Half of Pump.fun Launches
The data also challenges the idea that millions of tokens necessarily represent millions of independent creators experimenting with new projects.
Using Pump.fun launch records from mid-November through early July, Bitquery identified 1.26 million wallets responsible for 6.1 million launches. Three-quarters of those wallets created only one token.
At the opposite extreme, roughly one wallet in 200 launched more than 100 tokens. Together, that small group accounted for 48% of all launches in the dataset.
The busiest wallet averaged approximately one new token every four minutes.
Some of those wallets may belong to applications or automated launch services rather than individual developers, so the figures should not be interpreted as proof that a small number of people personally created half of Pump.fun’s coins. They do show, however, that token creation is highly concentrated and heavily automated.
That is consistent with a wider problem around interpreting blockchain activity. A separate Bitquery study recently found that two bot clusters generated 65% of Avalanche DEX volume, showing how raw transaction counts, token counts or trading volume can dramatically overstate the diversity of participants behind them.
451,402 Coins Skipped the Bonding Curve Entirely
The most concerning part of the research involves a separate category of tokens that never followed the normal Pump.fun graduation route.
Bitquery identified 451,402 brand-new coins through July 5 whose developers bypassed a launchpad bonding curve and instead created their own PumpSwap liquidity pools directly.
These so-called curve-skip pools could look remarkably similar to genuine graduated tokens. The median pool opened with about 85 SOL, roughly the same amount found in a normal Pump.fun graduation pool.
The crucial difference was control of the liquidity.
In a normal Pump.fun graduation, the migrated liquidity becomes part of the protocol’s canonical PumpSwap pool and cannot simply be withdrawn by the token creator. In a manually created pool, the wallet supplying the liquidity can retain the ability to remove it.
Bitquery found that 99% of the 451,402 curve-skip pools it tracked were eventually emptied, almost always by the developer that opened them.
The process was often extremely fast. The median pool survived just 12 minutes before the developer first withdrew funds, while 87% experienced their first withdrawal within an hour.
Most of the Apparent Buyers Were Connected to the Developer
Bitquery then investigated 160 of those pools more deeply by tracing where buyers had obtained their SOL.
The result complicates the appearance of active early trading even further.
According to the investigation, 95% of the SOL ultimately left behind by buyers came from wallets tied to the developer through direct transfers or a shared funding source. Most individual wallets were not connected under Bitquery’s test, but those outside wallets generally lost very little — about $1 each on average, or roughly 5 SOL per investigated pool collectively.
In other words, a pool could display a crowd of apparently independent buyers while much of the meaningful capital circulating through it originated from the same network surrounding the developer.
The distinction is particularly relevant in a market where Solana-based memecoins such as BONK have grown large enough to reach major centralized exchanges. The existence of successful outliers can make the long tail of newly created tokens appear more investable than its actual survival statistics suggest.
The Base Rate for a New Memecoin Is Now Measurable
For retail traders, the most useful part of the research may not be identifying individual rug pulls. It is establishing the base rate before a trade is placed.
A buyer opening Pump.fun and selecting a newly launched token is entering a universe where four out of five coins will never trade again after their first day and half will effectively be finished within about three minutes.
Only about 1.7% reach PumpSwap through the standard graduation process. Of the enormous number that do not graduate, very few maintain meaningful trading activity weeks later.
Those odds do not mean every new token is a scam. Most dead tokens may simply fail to attract sustained attention. The data measures trading survival, not fraudulent intent.
That distinction matters. Cryptocurrency networks routinely generate large headline numbers that need to be separated into their economic components, something also seen when examining what blockchain transaction volume actually tells investors.
But the base rate does change how early memecoin returns should be interpreted. A trader is not merely trying to identify which token might appreciate. They are first betting that the token will survive long enough to maintain a functioning market.
Graduation Is Useful, but It Is No Longer Enough
The research also suggests that one of the market’s traditional shortcuts — waiting for a token to graduate — needs refinement.
Graduated Pump.fun coins clearly survived at much higher rates than non-graduates. Yet a third of September’s graduations required three curve trades or fewer, meaning the milestone can sometimes be reached through a tiny number of sufficiently large purchases rather than a broad wave of demand.
Another Solana launchpad makes the problem even clearer. Bitquery found that 75% of tokens launched through Meteora’s Dynamic Bonding Curve graduated, with 35% graduating after a single trade. Yet graduated Meteora tokens were no more likely to remain active in their fifth week than those that did not graduate.
The word “graduated” therefore describes a technical event whose economic significance depends heavily on how a particular launchpad defines it.
Liquidity Origin May Matter More Than the Number of Buyers
The 99% figure for curve-skip pools is potentially the most actionable finding, but it also requires the most careful interpretation.
It does not mean 99% of Pump.fun coins are rug pulls. It does not even apply to normal Pump.fun graduation pools.
It refers specifically to 451,402 manually created PumpSwap pools for brand-new tokens that bypassed the bonding curve in Bitquery’s dataset through July 5. Almost all of those pools were eventually emptied by the wallet that created them.
That makes pool provenance a critical piece of due diligence. Two pools can begin with approximately the same 85 SOL and look similar to a trader scanning a new token, while having completely different protections around who controls the liquidity.
Thin liquidity can also turn what looks like a valuable token position into something far less realizable once selling begins. The recent collapse of liquidity around Solana-based SILV demonstrated the same underlying market principle in a very different type of token: quoted or nominal value matters little if the available liquidity cannot absorb sellers.
For new memecoins, Bitquery’s data pushes that lesson to an extreme. Token creation has become almost frictionless, but durable liquidity, independent buyers and sustained trading remain scarce.
Eleven and a half million new tokens sounds like explosive market growth. The more revealing statistic is what happened afterward: most disappeared from active trading almost immediately.
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape. You can reach out to him via his social media accounts:
Linkedin: https://www.linkedin.com/in/johan-shamshad-742851262/
X: https://x.com/Yasmine_FX
Investing: https://www.investing.com/members/contributors/279781574