pumpdata.funGet a free keyFor agents and builders →

One in a hundred pump.fun wallets makes more than half the trades, every week

From Jun 1 to Nov 1, 2026, the busiest 1% of wallets made between 55.8% and 67.0% of pump.fun's trades every week. In the week of Oct 26, that was 25,723 wallets and 46,551,417 of 83,431,269 trades, and 62.0% of the 2,572,337 wallets made 5 trades or fewer. Of the wallets that have traded 10 or more coins, 68.5% took out less SOL than they put in.

The long view · pumpdata.fun · published November 6, 2026 · 5 min read · markdown

Quick answer

In every week from Jun 1 to Nov 1, 2026, the busiest 1% of wallets made more than half of pump.fun's trades: between 55.8% and 67.0%. In the week of Oct 26 to Nov 1, 2,572,337 wallets made 83,431,269 trades of pump.fun coins. The busiest 25,723 of them made 46,551,417 of those trades, 55.8%. The busiest 10 alone made 7,995,499, or 9.6%. At the other end, 62.0% of the week's wallets made 5 trades or fewer, and 40.8% made 2 or fewer.

@18decimals quoted another tracker's week: 131 million trades by 4.1 million wallets, 69% of them from 1% of wallets, and 71% of wallets that closed at least 10 positions losing money. Our archive gives the same picture with our own counts, below.

Week by week

week (Mon, UTC)tradeswalletsbusiest 1%their share5 trades or fewer
Jun 136,486,0071,188,83211,88859.7%69.4%
Jun 837,371,0121,508,89715,08861.0%72.2%
Jun 1537,117,8431,665,39816,65360.6%73.1%
Jun 2237,724,7571,783,02317,83056.9%71.7%
Jun 2945,411,5041,714,05717,14059.1%69.5%
Jul 640,455,2321,678,66216,78658.8%71.4%
Jul 1340,812,8421,397,50113,97558.9%65.8%
Jul 2058,437,5192,062,32920,62356.1%63.2%
Jul 2768,165,7061,982,27419,82260.1%65.1%
Aug 372,026,5831,424,47314,24456.1%57.5%
Aug 1079,286,8421,425,58614,25556.2%57.6%
Aug 1792,275,1802,702,15027,02160.2%67.6%
Aug 24107,815,3382,868,51528,68560.0%67.4%
Aug 3197,106,6912,711,38027,11357.5%69.0%
Sep 7129,777,8312,772,68227,72660.3%66.5%
Sep 1497,868,6562,850,70328,50761.0%67.9%
Sep 21101,649,8382,611,56526,11559.2%64.1%
Sep 28114,783,4702,740,24127,40262.1%62.1%
Oct 5119,438,2712,697,64326,97664.8%64.5%
Oct 12112,226,5652,894,81628,94867.0%67.0%
Oct 19102,341,5193,256,97932,56962.7%67.4%
Oct 2683,431,2692,572,33725,72355.8%62.0%

pump.fun's BOOST mode arrived the week of Jul 20, and trading grew from 39,339,885 trades a week over the 7 weeks before it to 95,775,419 over the 15 weeks from it. The busiest 1%'s share barely moved: 59.3% before, 60.4% after. More wallets came, and they traded the way the earlier ones did: a few trades each, while a small group made most of the rest. The share peaked at 67.0% in the week of Oct 12; the last week counted was the lowest of the 22.

Who the busiest 1% are

The 25,723 busiest wallets of the week of Oct 26 each made 483 trades or more. Here they are by the class our wallet table gives them, as built on Nov 5:

classwalletstradesshare of all the week's trades
bot1,39219,560,22823.4%
none13,36313,485,53416.2%
winner6,3137,547,9249.0%
net_loser2,9444,274,2745.1%
supply_seller1,7111,683,4572.0%

1,392 wallets the table calls bots, 5.4% of the busiest 1%, made 23.4% of all the week's trades by themselves. The class rules are on our methodology page: a bot snipes new coins in their create slot or trades 1,500 or more times a day; a winner has taken out 10 SOL or more and 1% more than it put in over 20 or more coin-weeks; a net_loser has put in 50 SOL more than it took out; a supply_seller has taken out 1.5 times what it put in, 100 SOL or more; none is a record no rule fits yet, most often a young wallet.

The same wallets fire most of the trades that fail. The week had 167,235,553 failed trades, about 2 for every landed one, and 82.1% of them came from the week's busiest 1%.

Do most of them lose?

Our wallet table holds every wallet's SOL in and out on pump.fun since pump.fun's first day. Of the 7,722,148 wallets that have traded 10 or more coins, 68.5% took out less SOL than they put in. That is close to the tracker's 71%, counted a different way: theirs is closed positions, ours is SOL in against SOL out over every coin a wallet touched, with an unsold balance counted as nothing. The share falls a little the more coins a wallet has traded: 66.7% of wallets with 50 or more coins, 64.4% with 100 or more. Over all 98,653,307 wallets in the table, it is 57.2%.

What we counted

Trades are the archive's landed trades of pump.fun coins: every bonding-curve trade, and PumpSwap trades in each coin's own pool, the SOL pool its migration made. Pools a coin's traders opened later, and other tokens' pools, are left out, which is why our week has fewer trades and wallets than the tracker's. A wallet is the trade's signer. Weeks run Monday to Sunday, UTC. The busiest 1% is the top hundredth of each week's wallets by its own trade count, so the wallets change from week to week. Failed trades are both venues' failed trade instructions in the week's slots. The classes and SOL in and out come from our wallet table as of Nov 5, 2026. These are counts of what happened on chain, not a judgement of anyone, and none of this is advice.

Ask about a wallet yourself

Every wallet's record, its class and why, and its SOL in and out, is one call: POST /wallets for up to 50 addresses, or /wallet/{address} for one wallet's page. The wallets that have taken out the most are on the leaderboard.

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.