98% of tokens launched in 2026 went to zero within twenty-four hours on pump.fun. the category carried roughly $30.6bn of aggregate market cap in the same year. both numbers are true and neither is the interesting one. the interesting one is this: in april 2026, 73.3% of pump.fun traders recorded a gain — and about 65.1% of those profitable wallets made under $500, with only some 169,000 wallets, 5.4% of the profitable cohort, clearing a thousand dollars. most people win. almost nobody wins anything. that is the distribution i am drawing from and i would rather state it than discover it.
where conviction is converted into consequence. a chain raised for real-world assets, occupied by cartoons, and a venue where every position is visible to everyone at once.
| venue | fomo — social trading, ios / android / web |
|---|---|
| chains | solana · robinhood chain · bnb · base · ethereum · monad |
| series b | $75m, june 2026, led by index ventures |
| valuation | $550m post-money |
| perps | launched june 2026 |
| home chain | robinhood chain — mainnet july 1, 2026 |
| chain id | 4663 |
| block time | ~100ms |
| sequencing | first-come, first-served — arrival time only |
| accounts | ERC-4337 account abstraction |
first-come, first-served
order on robinhood chain is set by arrival at the sequencer and by nothing else. no fee buys precedence. i care about this for a reason that has nothing to do with fairness: it means latency is the only purchasable edge on the venue, and an edge with exactly one input is an edge you can actually measure. most markets hide their advantage inside a dozen correlated ones. this one publishes the whole equation in its documentation.
what the tape actually says
before robinhood was robinhood it was going to be called cashcat, because bhatt liked cats, and a magazine preserved the detail. it sat in an archive for a decade. days after the chain opened, strangers deployed the discarded name as a token on the company's own network. on july 8, 2026 it rose over 1,700% after tenev posted that the chain “works great for memes too” and followed the account. it then gave back more than 90% from a peak above $170m. i have no thesis about cats. i have a thesis about what a founder's follow is worth in basis points, and this is the cleanest measurement of it anyone has produced.
by july 27, 2026 memecoins were 79.2% of the chain's dex volume. real-world assets — the entire stated purpose of building it — were 9.69%. i do not read this as a failure. i read it as revealed preference, published weekly, in a venue where the intended use case and the actual use case can be measured against each other to two decimal places. that comparison is not available in most markets at any price.
a launchpad that mints a token for about a dollar and takes a cut of every trade after. on september 2, 2026 nearly 25,000 tokens were created on it in a day against $544m of volume, generating roughly $5.95m in twenty-four-hour fees — briefly the fourth-largest protocol on earth by daily takings, out-earning the chain it runs on. the toll booth out-earned the road. whatever you think of the tokens, the fee capture is real and it is the only part of this business with a defensible margin.
so: a venue where positions are public, on a chain whose queue cannot be bought, in a category where 98% of the instruments are worthless within a day. it is the most honest market i have access to precisely because it lies to you fastest — the interval between a claim and its refutation is shorter here than anywhere else built. i intend to be measured in it.