On-chain investigationTokenized stocksLiquidity pools

Tokenized stock LPs earned $7.3M in fees and lost $9.7M to impermanent loss. Only two Solana venues paid without rewards

Every swap of a stock token on a DEX needs a liquidity provider on the other side. We tracked what those providers earned and gave back on Base, Solana, BNB Chain and Robinhood Chain.

By · Blockchain IntelligenceOn-chain figures verified November 13, 202611 min read
At a glance
From 14 October to 10 November 2026, traders swapped $7.8 billion of stock tokens through 656 DEX pools on five chains. The liquidity providers behind those pools collected $7.30M in fees. Impermanent loss, what price moves cost them against simply holding the same tokens, came to $9.65M. AERO rewards on Base added $1.67M, so they finished $0.68M down, a loss of 6% per year on $163 million of liquidity. Only two venues paid without rewards: Orca and Raydium on Solana. The biggest venue, Aero on Base, lost money on trading and made it back in AERO.
$7.30M
Fees liquidity providers kept, 14 Oct to 10 Nov
−$9.65M
Impermanent loss: lost to price moves against holding the tokens
−6%
Net result per year, after $1.67M of AERO rewards
2
Venues that paid without rewards: Orca and Raydium on Solana
Key findingsEach line has its own share link
  1. −6%

    Liquidity providers in 656 tokenized stock pools kept $7.30M in fees and lost $9.65M to impermanent loss in four weeks. With AERO rewards they ended $0.68M down, a loss of 6% per year.

  2. 2 venues

    Only Orca (+24% per year) and Raydium (+11%) on Solana paid their liquidity providers without rewards. Meteora, on the same chain, lost 37%.

  3. −169%

    Aero pools on Base, where Coinbase's stock tokens trade, lost 169% per year on trading. AERO rewards turned that into +37%.

  4. −$10.7

    Pre-market trades cost pools $10.7 per $10,000 traded. Overnight trades paid them $2.9 and weekend trades $2.0.

  5. 60 s

    Against the Hyperliquid perp, Aero and Robinhood Chain Uniswap pools are below zero a minute after a trade. Raydium pools still keep $7.5 per $10,000.

  6. 17 of 20

    In a replay of 7-day positions, moving a range to follow the price lost money in 17 of 20 cases. A ±10% range left alone made money in three of four pools.

01 · The other side of the tradeSomeone has to sell you that NVIDIA token at 3am

You can buy an NVIDIA token on a DEX at 3am on a Sunday. The stock market is shut. The trade still goes through in seconds.

The other side of that trade is a pool of tokens that ordinary wallets put up. Anyone can add money to a pool and collect a fee each time somebody trades against it. Crypto calls these people liquidity providers, or LPs. Tokenized stocks now trade this way on Solana, Base, BNB Chain and Robinhood Chain, and a little on Ethereum. Coinbase, Robinhood, Binance and xStocks all have tokens in these pools, along with smaller issuers.

So do the LPs make money? Bitquery records every swap on these chains, so we measured it. We took four weeks of trades, from mid-October to 10 November, in the 656 open pools whose size we could track. Together they handled $7.8 billion of stock-token trading, most of the market. Our October report puts all DEX trading in tokenized stocks at $10.3 billion that month.

The short answer is no. LPs earned $7.30M in fees and gave back $9.65M as prices moved against them. AERO rewards on Base cut the damage to $0.68M, a loss of 6% per year. Two venues on Solana paid. Everywhere else, LPs either lost money or lived on rewards. The two venues that paid charged fees high enough to outlast the traders who see the price move first.

02 · How LPs earn and loseA fee on every trade, and a price gap on some

A liquidity pool is a pot of two tokens, for example NVIDIA tokens and USDC, that trades at a price set by its own balance. An LP deposits both tokens and owns a slice of the pot. Every trade pays a small fee into it, usually a few hundredths of a percent and at most a quarter of a percent.

The cost comes from the fact that a stock token copies a share that trades somewhere bigger. When NVIDIA jumps on Nasdaq, on a perp exchange or on another pool, this pool keeps quoting the old price for a moment. An arbitrage bot buys the cheap tokens and sells them where the price has already moved. The pool collects the fee and gives up the gap.

That gap is what DeFi calls impermanent loss. It is the difference between what the LP's slice is worth and what the same tokens would be worth left in a wallet. "Impermanent" is generous: once the LP withdraws, the loss is permanent. An LP makes money only when fees beat the gap.

Every trade pays a fee. Some trades also take a gaphow a stock-token pool earns and loses
WHERE A LIQUIDITY PROVIDER EARNS, AND WHERE THE MONEY LEAKS OUTThe stock movesNVIDIA jumps 1% on news.Bigger venues repricewithin seconds.The pool lagsThe pool still sellsNVIDIA tokens at theold, cheaper price.An arb trader buysA bot buys the cheaptokens from the pool andsells them elsewhere.The pool keeps the feeEvery trade pays it: 0.02% to 0.25% of thetrade on 80% of the volume we measured.The pool gives up the gapIt sold tokens below the new price. When thegap beats the fee, providers lose money.
A stock token tracks a share that trades on bigger markets. When the share moves first, the pool sells at a stale price until someone trades it back in line. The fee is the provider's income; the gap is the provider's cost.

Stock tokens make this worse. The real price lives off-chain, on markets that open and close and react to news the pool cannot see. Anyone who sees a move first, on a perp or in the pre-market, can trade against the pool before it catches up.

03 · The scoreboard$7.30M in fees, $9.65M back to the market

We valued every swap against the stock token's price on all other pools at the end of that day. Fees count as income. Protocol fees and the share some DEXs send to their token voters do not, because LPs never see that money. The chart adds it all up day by day.

Fees came in every day. Price moves took morerunning totals for liquidity providers, 14 October to 10 November 2026
RUNNING TOTALS FOR LIQUIDITY PROVIDERS IN 656 POOLS, $ MILLION-10-8-6-4-20+2+4+6+814 Oct21 Oct28 Oct04 Nov10 NovFees earned+$7.30MAERO rewards+$1.67MNet result−$0.68MImpermanent loss−$9.65M
Each line adds up one part of what liquidity providers earned or lost, day by day, across the 656 pools we measured. "Impermanent loss" is what they gave up against simply holding the same tokens. AERO rewards go only to Base pools.

Fees came in every day, but the price-move loss grew faster and ended bigger. Before rewards, LPs were down 20% per year on the money they put in. For every $10k traded, they kept about $9 in fees and gave back about $12.

Solana LPs came out about flat. Robinhood Chain and BNB Chain LPs lost money, and Base LPs ended ahead only because of AERO rewards. Fewer than half of the pools, 43%, came out ahead.

04 · Venue by venueTwo Solana venues paid. The rest lost before rewards

The same stock token often trades on several venues at once, and the venue made the biggest difference. Orca Whirlpool and Raydium CLMM, two Solana DEXs, paid their LPs +24% and +11% per year. Every other venue with $60 million or more in trades lost money before rewards.

Only two venues paid without rewardsresult for liquidity providers by venue, as a yearly rate, 14 October to 10 November 2026
RESULT FOR LIQUIDITY PROVIDERS, % PER YEAR ON POOL VALUEBefore rewardsAfter AERO rewards-80%-60%-40%-20%0+20%+40%+60%Orca WhirlpoolSolana · $392M traded+24%Raydium CLMMSolana · $968M traded+11%Uniswap v4BNB Chain · $130M traded−3%AlgebraRobinhood Chain · $61M traded−9%Uniswap v4Robinhood Chain · $558M traded−12%Uniswap v3BNB Chain · $120M traded−12%Uniswap v3Robinhood Chain · $1,219M traded−18%Byreal CLMMSolana · $63M traded−19%PancakeSwap v3BNB Chain · $931M traded−19%Meteora DLMMSolana · $404M traded−37%Aerodrome (old pools)Base · $745M traded−52%+16%Aero Slipstream V3Base · $2,112M traded−169%+37%
Venues with $60 million or more traded. Algebra on Robinhood Chain is the smallest. Result is fees minus what price moves cost, against holding the same tokens. The Aero bar stops at −80%: before AERO rewards its pools lost 169% per year over the 21 days they were live. Only Base pools earn AERO.

A high fee helps but does not decide it. Raydium and Orca LPs kept more in fees than they lost to price moves. Byreal CLMM LPs kept more per trade than Orca's and still lost 19% per year, because fast traders picked off their pools harder. Meteora DLMM, also on Solana, lost 37% per year.

High fees help, but only if the loss stays smallerfees kept and what price moves cost for each $10,000 traded, by venue
PER $10,000 TRADED: FEES KEPT AGAINST THE LOSS TO PRICE MOVES, IN DOLLARSFees keptImpermanent loss$30$20$10$0$10$20$30Orca WhirlpoolSolana$12$10net +$2.7Raydium CLMMSolana$20$17net +$3.0Uniswap v4BNB Chain$5$5net −$0.2AlgebraRobinhood Chain$1$2net −$0.3Uniswap v4Robinhood Chain$15$19net −$3.7Uniswap v3BNB Chain$8$9net −$1.4Uniswap v3Robinhood Chain$11$15net −$3.8Byreal CLMMSolana$13$23net −$9.6PancakeSwap v3BNB Chain$15$19net −$4.1Meteora DLMMSolana$11$16net −$5.9Aerodrome (old pools)Base$3$7net −$3.1Aero Slipstream V3Base$2$7net −$5.3
One basis point is $1 on every $10,000 traded. "Fees kept" is after protocol fees and the share Aero and Aerodrome send to voters. Aero pools keep the least because most of their fees go to voters.

On Robinhood Chain, Uniswap v3 pools handled $1.2 billion and lost 18% per year. Most of those pools also send part of every fee to the Uniswap protocol, money the LPs never receive. PancakeSwap v3 LPs on BNB Chain kept more per trade and still lost 19% per year.

05 · Base runs on AEROCoinbase's stock tokens trade in pools paid in AERO

Coinbase's stock tokens trade mostly on Base, in Aero Slipstream V3 pools. Aero is the DEX formed when Aerodrome merged with Velodrome, and its pools went live on 21 October. Most LPs there stake their positions in Aero's gauges, its reward contracts. Staked liquidity gives its swap fees to AERO voters and earns newly issued AERO instead.

Voters took 69% of the swap fees on these pools. Price moves then cost LPs more than the fees they kept, and AERO rewards turned the total into a small gain.

On Base, AERO rewards pay the billAero Slipstream V3 pools, from launch on 21 October to 10 November 2026
AERO SLIPSTREAM V3 POOLS ON BASE, 21 OCT TO 10 NOV, $ MILLION-1.0-0.50+0.5+1.0+$1.15MSwap fees earned−$0.80MSent to voters−$1.47MImpermanent loss+$1.36MAERO rewards+$0.25MNet result
Most LPs in these pools stake their positions in Aero's gauges. Staked liquidity hands its swap fees to AERO voters and earns AERO instead. We value AERO at each day's average price; the token fell from $0.89 to $0.55 over the window.

As a yearly rate over the weeks the pools were live, Aero LPs lost 169% on trading and made +37% with AERO. Across all Coinbase stock-token pools, LPs lost money on trading and made a small profit once AERO came in.

That profit leans on the price of AERO. We valued each day's rewards at that day's average price, and AERO fell from $0.89 to $0.55 over the four weeks. An LP who held the rewards lost part of them to that slide. If AERO were worth a fifth less, Base LPs would be about flat. Our earlier look at Coinbase's tokens found that one group of wallets did more than a third of all their trading.

06 · When LPs losePre-market trades cost the most

Tokenized stocks trade around the clock, so we split every trade by the hour it happened in New York. Night and weekend trades, when Wall Street is shut, left LPs ahead. Trades in the pre-market and after hours cost them the most. Those are the hours when US companies usually publish earnings and the real stock trades thinly on Nasdaq.

Night and weekend trades paid. Pre-market trades cost the mostresult per $10,000 traded, by session in New York time, at the day's closing price
WHAT POOLS KEPT PER $10,000 TRADED, BY WHEN THE TRADE HAPPENED, MEASURED AT THE DAY'S END−$12−$8−$4$0+$4+$8+$12Overnight, weekdays8pm to 4am New York · $1.8B traded+$2.9WeekendFriday 8pm to Sunday 8pm · $1.3B traded+$2.0Regular hours9:30am to 4pm · $2.6B traded−$4.2After hours4pm to 8pm · $0.9B traded−$7.4Pre-market4am to 9:30am · $1.3B traded−$10.7
We value each trade at the stock token's price on other pools at the end of that day. Fees count as income; protocol fees and voters' shares do not.

Pre-market trades cost pools about $11 for every $10k traded, and after-hours trades about $7. Together those two sessions cost LPs about $2 million. Our weekend study found that stock tokens price in most of Monday's move only once the US market wakes up, which fits this pattern.

The second test asks how fast the edge goes. For nine stocks with a perp on Hyperliquid, we compared each pool trade with the perp's price at the moment of the trade and again one minute later. The perp usually reacts first, so it shows whether the trader on the other side of the pool knew something.

A minute is enough to strip the edge on most EVM poolswhat pools kept per $10,000 traded, at the trade and 60 seconds later, against the Hyperliquid perp
WHAT POOLS KEPT PER $10,000 TRADED, AGAINST THE HYPERLIQUID PERP PRICEAt the trade60 seconds later−$2$0+$2+$4+$6+$8+$10Raydium CLMMSolana+9.3 → +7.5PancakeSwap v3BNB Chain+5.3 → +3.7Orca WhirlpoolSolana+3.1 → +2.3Uniswap v3BNB Chain+1.9 → +0.3Uniswap v3Robinhood Chain+1.7 → −0.3Aerodrome (old pools)Base+1.3 → −0.4Uniswap v4Robinhood Chain+3.0 → −0.7Aero Slipstream V3Base+1.0 → −0.9
Nine stocks with a Hyperliquid perp and pools priced in stablecoins. We remove the token's usual gap to the perp first. Meteora pools on Solana sit off the scale: they lost about $21 per $10,000 at every step.

On Aero pools, LPs were slightly ahead at the moment of the trade and slightly behind a minute later. Uniswap pools on Robinhood Chain followed the same path. Fast traders take the whole edge on these low-fee pools within a minute. Raydium pools still kept more than $7 per $10k after the minute, because their fees are high enough to absorb the move. Our piece on a Saturday crash in Hyperliquid's stock perps shows how far those perps can run while Wall Street is closed.

07 · Range choiceWide and patient beat tight and busy

Most of these pools let an LP pick a price range. A tight range earns more fees while the price stays inside it, and earns nothing once the price leaves. We replayed small positions in four of the largest EVM pools, one opened every day for three weeks and held for a week, with different ranges.

Chasing the price turned winners into losersreplay of positions opened each day from 14 October to 3 November and held 7 days, median result as a yearly rate
REPLAYED 7-DAY POSITIONS: MEDIAN RESULT, % PER YEARFull range±10%, left alone±5%, left alone±5%, moved with price-150%-100%-50%0+50%+100%SpaceX / USDGUniswap v3, Robinhood Chain+4%+71%+65%−87%MicroStrategy / USDCAerodrome, Base+4%+86%+70%−85%NVIDIA / USDTPancakeSwap v3, BNB Chain+0%+4%+7%+1%SanDisk / USDCAero Slipstream V3, Base−9%−174%−252%−306%
Each replayed position is small enough not to move the pool and earns its share of fees while the price sits inside its range. "Moved with price" re-centers the range whenever the price leaves it and pays the swap fee to rebalance. Bars stop at −150%.

Chasing the price lost money. Re-centering the range every time the price left it lost in 17 of 20 pool and range cases, because each move locks in the loss and pays a swap fee to rebalance. On SpaceX on Robinhood Chain, a ±5% range that followed the price lost 87% per year, while a ±10% range left alone made 71%.

A range left alone works only when the stock goes nowhere. A ±10% range made money in three of the four pools. In the SanDisk pool on Aero, the token fell 39% in four weeks, and every range we tested, up to ±10%, lost money.

08 · Who the LPs areA handful of wallets, moving fast

We looked at who adds liquidity to the 30 pools that paid the most fees. In the median EVM pool among them, the five largest wallets put in 60% of all the money added.

A handful of wallets supply most of the money, and on Base they move it within hourstop 5 wallets' share of the dollars added to each pool, and the median time a closed position stayed open
THE 20 EVM POOLS THAT PAID THE MOST FEES: TOP 5 WALLETS' SHARE OF MONEY ADDED, AND HOW LONG POSITIONS LASTEDBaseRobinhood ChainBNB ChainMEDIAN TIME HELD0%25%50%75%100%MSTR · Aerodrome, old98%6 hoursSNDK · Aero90%2 hoursSPCX · Aero88%8 hoursMSTR · Aero84%5 hoursAMZN · Aero80%4 hoursMETA · Aero76%5 hoursMSTR · Aero66%2 hoursGOOGL · Aero63%5 hoursMSTR · Uniswap v490%16 hoursMETA · Uniswap v468%24 hoursCRCL · Uniswap v357%3 daysSPCX · Uniswap v353%14 hoursTTWO · Uniswap v343%3 hoursUSO · Uniswap v318%5 daysNVDA · PancakeSwap v353%32 hoursSPCX · PancakeSwap v350%40 hoursFXI · PancakeSwap v349%19 hoursBNC · PancakeSwap v347%20 hoursBNC · PancakeSwap v330%under 1 hourBNC · PancakeSwap v315%1 hour
A wallet is the address that sent the deposit. Money added counts every deposit, so a manager that pulls and re-adds the same money every few hours counts each time. Uniswap v4 pools record no token amounts, so their share uses liquidity units instead.

On Base, positions are short. The median closed position in an Aero pool lasted a few hours. The SanDisk pool alone had 874 LP wallets, but its top five added 90% of the money, and LPs added and removed hundreds of positions within a single block. That pattern fits active managers and bots that keep a tight range around the price and farm AERO. On Robinhood Chain and BNB Chain, the top five wallets held a smaller share in most pools, and the median position lasted anywhere from under an hour to five days.

09 · Is there an opportunity?Yes, in a few places, for a few kinds of LP

On average, providing liquidity to tokenized stocks lost money in these four weeks. The data also shows where it did not.

  • High-fee Solana pools paid. Orca and Raydium pools returned a profit, and their fees covered the price gap even a minute after a trade.
  • Rewards can outrun the loss. Aero LPs lost money on trading and made it back in AERO. That only works while the reward token holds its price and the rewards keep coming.
  • Wide ranges left alone held up. In the replay, a ±10% range beat a tight range that followed the price on all four pools.
  • The quiet hours paid. Overnight and weekend flow made LPs money. Pre-market and after-hours flow cost them. A pool cannot pick its traders, but a manager can widen ranges or step back around those hours.

Where none of that applies, a stock-token LP is selling cheap options on the stock to faster traders. Coins priced in stock tokens push more flow into the same pools. 39% of Solana and 44% of BNB Chain stock-token swap volume ran through a transaction that also traded such a coin. Our Pump.fun stock-token study covers that market. We did not measure how much of the loss that flow causes.

10 · How we measuredWhat we compared

We took every swap in tokenized stock pools over the four weeks and valued it at the end of that day, using the token's price on all other pools. Fees went in as income; protocol fees and voters' shares came out. We tracked each pool's size every day to turn dollar results into a yearly rate.

The trade data, pool sizes and wallet records come from Bitquery's data for Base, Solana, BNB Chain and Robinhood Chain. The full history of each chain is also sold as files. Our market count lists the stocks and issuers we covered, and our NVIDIA comparison shows how closely their tokens track each other.

The net result is a small difference between two large numbers, so read it as a direction rather than a precise rate. Four weeks is one stretch of market, including one weekend of big news, and a yearly rate here describes that stretch. It is not a forecast.

FAQ

Do liquidity providers make money on tokenized stocks?

On average, no. In the four weeks we measured, LPs in tokenized stock pools kept $7.30M in fees, and price swings cost them more. Even with AERO rewards on Base, they lost 6% per year.

Is there impermanent loss in tokenized stock pools?

Yes, and it was larger than the fees on most venues. The stock's real price moves first on bigger markets, and arbitrage traders buy from the pool before it catches up. That gap cost LPs about $12 per $10k traded, against about $9 kept in fees.

Which tokenized stock pools paid their liquidity providers?

Orca Whirlpool and Raydium CLMM pools on Solana. Aero pools on Base paid only with AERO rewards included.

What does a crypto liquidity provider earn in a tokenized stock pool?

A share of the fee on every swap, plus any rewards the venue pays. After protocol fees, LPs kept about $9 per $10k traded on average.

When do tokenized stock pools lose the most?

In the pre-market and after hours in New York, when US companies usually publish earnings and the real stock trades thinly. Overnight and weekend trades left pools ahead.

Limits on these figures

The window is 14 October to 10 November 2026. Results cover the 656 open pools whose daily size we could track, which handled $7.8 billion of trading. We leave out market-maker venues that quote from their own capital, because that capital is not on chain.

We value each swap at the stock token's price on other pools at the end of that day, so the result resets daily and does not follow any single LP's entries and exits. The day-end price comes from the last minute of trading on other pools, which is thin for some stocks.

We read fees on EVM chains from each swap. We take out protocol fees for PancakeSwap, Uniswap v3 and Robinhood Chain's Uniswap v4 pools, and Aero's voter share. We did not check protocol fees on BNB Chain and Base Uniswap v4 pools or Algebra pools; together they earned about $0.1 million in fees. We read Solana fee settings on 12 November and left out Meteora's variable fee, so its fees are a floor.

We count AERO rewards when Aero credits them to each pool's gauge, at that day's average AERO price. We do not count gas, hedging costs or taxes.

The range replay uses four EVM pools and small positions that do not move the pool. The provider counts cover the 30 pools that paid the most fees; a wallet is the address that sent the deposit.

Legal disclaimer

This article is provided for informational and educational purposes only and reflects analysis of publicly available on-chain data as of the dates indicated. It does not constitute legal, financial, compliance, or investment advice, and nothing in it is a recommendation to buy, sell, hold or provide liquidity for any token or asset.

The findings describe swaps, pool balances, fees and rewards observed on public blockchains from 14 October to 10 November 2026. They may be incomplete or incorrect and may be revised as more data becomes available.

References to Coinbase, Robinhood, Binance, xStocks, Aero, Orca, Raydium, Meteora, Uniswap, PancakeSwap, Hyperliquid or any listed company describe what the public record shows about their tokens and pools. Nothing here asserts that any party acted unlawfully.

Nothing herein should be relied upon as a definitive determination of fact. Readers should conduct their own independent verification before taking any action. The authors and publisher accept no liability for any loss or damage arising from reliance on this material. All trademarks and company names are the property of their respective owners.

Reported by Gaurav Agarwal for Bitquery Research, with AI tools; every figure was checked against the raw data.

Run it yourself

Check a pool before you add liquidity

The swaps, pool sizes and rewards in this piece come from Bitquery's data for Base, Solana, BNB Chain and Robinhood Chain. The Bitquery MCP server puts that data behind an AI assistant, so you can ask what a pool earned in fees, who trades against it or how its price compares with other venues, without writing the query yourself.

Pull every swap in a poolCompare a token's price across venuesList the wallets adding liquidityTrack a pool's size over time
Explore Bitquery MCP →Figures measured on swaps from 14 October to 10 November 2026 in Bitquery's data for Base, Solana, BNB Chain and Robinhood Chain.

Run the next investigation in minutes

Ask any question of the blockchain. Bitquery's AI Investigations agent traces funds, attributes wallets, and writes auditable reports across every major chain. Prefer to do it yourself? Connect the Bitquery MCP server to your AI assistant, pull the raw datasets, or query the chain directly with the Bitquery API.

Explore AI Investigations →Get a free API key