CryptoSlate
Only 26% of $752.5 million in large Pump.fun wallet gains passed Bitquery's checks for conventional market trading, according to an Oct. 22 investigation covering 3,170 wallet-token positions with reported gains of at least $100,000. Bitquery found 51% of the gains involved wallets selling tokens received from other addresses, 13% showed potentially coordinated trading patterns, and the remaining 10% could not be classified.
Why it matters
Copy-trading platforms like GMGN execute purchases based on buying transactions made by monitored wallets. A wallet that receives tokens through transfers can display large profits with no acquisition cost on record, but followers get no purchase signal to enter at a comparable price. Bitquery traced 1,104 of 1,579 transfer-based winning positions to addresses associated with the token's creator.
The practice is accelerating. Transferred-token sales accounted for seven of every 10 large wallet wins in August, after staying below 100 cases monthly until July. Among the 25 largest reported gains, 15 met Bitquery's criteria for wash loops, where concentrated buying coincided with direct financial links between wallets. One Arena wallet spent $4,223 before recording roughly $4.92 million in sales, with a linked address accounting for 99% of buying during the holding period.
Market impact
Wallet leaderboards, the core discovery tool for copy traders, may systematically surface addresses whose histories offer little indication of returns available to outsiders. GMGN's own documentation flags "Didn't buy" and "Sold > Bought" patterns, but its published materials do not establish whether profitability rankings account for transferred inventory.
Genuine performers exist in the sample, including a SOULGUY position that generated $3.3 million across 3,632 trades, but replicating such performance requires frequent execution at competitive prices, with slippage, fees and filters eroding returns further. Bitquery did not measure actual copy-trader results, and 2,943 additional six-figure winning positions closed after its August cutoff, so the true replicable share of headline Pump.fun profitability remains unquantified.
Frequently asked questions
-
Why can't copy traders replicate big Pump.fun wallet gains?
Many winning wallets acquired tokens through transfers rather than market purchases. Copy-trading bots execute only observable buys, so followers receive no entry signal and cannot match the displayed profits.
-
What share of Pump.fun profits came from conventional trading?
Bitquery found only 26% of $752.5 million in large wallet gains passed its checks for conventional market trading, while 51% involved selling tokens received from other addresses.
-
How are token creators inflating wallet profit rankings?
Creators can buy tokens in one wallet, transfer them to another, and sell there after a price increase. The receiving address shows large profits with no recorded acquisition cost, distorting leaderboard rankings.
-
What are wash loops in the Pump.fun study?
Fifteen of the 25 largest reported gains met Bitquery's wash-loop criteria, where concentrated buying activity coincided with direct financial links between wallets, such as a linked address accounting for 99% of buying during a position's holding period.
-
Did Bitquery measure actual copy-trader returns?
No. Bitquery did not examine copy traders' actual performance, and 2,943 additional six-figure winning positions closed after its August cutoff, so the replicable share of headline profits remains unmeasured.
Whale Alert
Crypto Rank News
Lookonchain
Glassnode