Correlated Pairs: How AMMs Win the Biggest Markets

Written by @haydenzadams, Founder of @Uniswap.

I’ve spent 9 years working on the frontier of DeFi. It’s a fascinating surface area with infinite depth and the capability to transform capital markets.
I’ve always believed in the massive potential of the AMM, but for the past decade, one big question nagged at me. Can this new market structure really become the core engine for all financial markets?
It’s taken years of evolution and growth, but there’s an increasingly clear path to AMM global dominance forming. The best way to explain it starts in 1976.
Tokenization changes who makes markets
The index fund turns 50 this month. When Jack Bogle launched it in 1976, he hoped to raise $150 million. He collected $11.3 million. Rivals called it "Bogle's folly" and printed posters calling index funds un-American. A fund that made no decisions, they argued, could never beat professionals paid to make them. Today the majority of American fund assets sit in passive vehicles.
I've been thinking about this lately, because tokenization’s folly moment is ending. The SEC has approved Nasdaq and NYSE to trade tokenized shares. DTCC, which settles nearly every security in America, ran a live trial of tokenized trading in July. Nearly all this activity is described the same way: tokenization as an infrastructure upgrade.
Same markets, faster, cheaper, always on. This is all true, but I think the upgrade framing buries the bigger story. Tokenization makes markets programmable: changing what markets exist, who makes them, and what those markets trade against.
In 2018 I built Uniswap, a protocol that automates market making. Anyone can deposit two assets into a shared pool and earn a fee on every trade, while prices adjust along a curve as people buy and sell. Uniswap has run autonomously since the day it launched, settled more than $4.6 trillion in volume, and helped push decentralized exchanges from under 1% of centralized spot volume to over 20%.
As AMMs like Uniswap have grown, their liquidity has organized into a pattern most of finance hasn’t noticed yet: correlated pairs.
The easiest wins
To win at everything, you have to start by winning something. AMMs found product-market fit in the long tail, where most assets couldn't get a professional market maker's attention at all. On Uniswap, anyone could create a market in one transaction, with issuers and early supporters as the first LPs.
Stable pairs came next. On a pair like USDC/USDT, a good passive strategy gets close enough to optimal that a lower cost of capital more than covers the gap. This is why professional trading firms don’t bother market making these stable swaps today. They’re being undercut by passive AMMs.
Big margins, no competition
Traditional financial markets fully belong to market-making firms. They bundle capital, trading strategy, execution technology, settlement, and distribution into one vertically integrated business. The architecture evolved for good reasons. Assets lived in separate systems, settlement was slow, and someone had to perform every function, so one firm performed them all.
With enough scale, all those fixed costs pay for themselves. Citadel Securities handles roughly 25% of American equity volume and produced a record $12.2 billion in net trading revenue last year on roughly $21 billion of trading capital.
Most people read those numbers as proof the system works. I read them as entrenchment.
Breaking the bundle
Blockchains create competition at each layer, breaking the bundle apart. Execution happens through code. Custody and settlement are shared services anyone can plug into. What once required proprietary infrastructure is now open source software.
With AMMs, capital is the scarce input and the edge belongs to whoever holds inventory most cheaply. A trading firm needs high returns to justify its overhead, so an LP willing to accept less undercuts it. Most market makers hedge away all price exposure, and hedging costs money, so an investor who already holds the assets takes that exposure for free. And asset issuers have a negative cost of capital, since asset issuers typically have to pay professional market makers to market make on their new assets.
Put simply, DeFi and AMMs lower the barrier to making markets, opening the space up to many new participants. Their edge can come from many different sources such as a lower cost of capital, desiring the inventory exposure professional firms would typically hedge, or even being the issuer themselves.
But everything rests on one question: can automated strategies perform well enough for this to hold?
Liquidity follows correlation
Recently, I was on a call with one of the largest institutions in finance. They asked me what base pairs were most common in DeFi. I explained that Ethereum-based assets tend to trade against ETH, Solana assets against SOL, and stablecoins pair against each other, with a small number of highly liquid pairs bridging between these clusters.

No one designed that. It emerged organically, in part because LPs do best when the assets they hold move together. Correlation means less inventory risk for liquidity providers, deepening liquidity. As assets tokenize, the biggest markets in the world will reorganize the same way.
They can't today. Traditional markets settle overwhelmingly in dollars out of necessity. Assets live in siloed systems, and fiat rails like SWIFT and Fedwire are the glue that holds everything together. But blockchains are a far more expressive glue. Tokenize the assets, and they share a settlement layer, so any asset can trade directly against any other.
NVDA/USD can become NVDA/SPY, with SPY/USD as the bridge back to dollars. Oil companies can trade against an oil ETF or tokenized oil. Private credit can trade against tokenized Treasury funds. Tokenization also enables markets that span different types of assets, which is extremely difficult, if not infeasible, with TradFi infrastructure.
Delta neutrality is inefficiency
Traditional market-making firms generally try to be “delta neutral,” which is fancy traderspeak for denominating in dollars and wanting to minimize any non-dollar risk. When making a volatile asset they'll pay money to reduce their non-dollar risk (aka hedging), usually through options. This is one of the higher cost aspects of traditional market making.
Pairing assets into lower volatility “correlated pairs” connected by a few higher volatility “bridge pairs” brings numerous efficiency unlocks, but the most important one is that market making is cheaper and more efficient if the people market making on the assets actually want to hold the underlying assets.
And the more correlated a pair, the smaller any gap between today’s passive AMM strategies and the most sophisticated active strategies, making it easier to “undercut” them with this lower inventory cost.
To make it concrete, if you're long NVIDIA, you’re probably also long SPY, and the gap in efficiency between passive AMMs and active strategies is much lower for NVIDIA/SPY than for NVIDA/USD.
Connected liquidity
If stocks trade against SPY instead, every trade that starts or ends in dollars routes through the same pair: SPY/USD. These bridge pairs still require sophistication. However, there are much fewer of them, and they carry so much flow that professional attention is worth it.
DeFi already runs this way. ETH/USDC is one of the deepest markets onchain because every cluster routes through it. Passive LPs supply the correlated pairs, while active LPs compete over the bridge pairs.
Investors can still buy and sell everything in dollars, since routing across pools is automatic. And liquidity will concentrate where risk is lowest, not where legacy plumbing requires it to sit. That pushes the deepest markets into correlated pairs, the ground where AMMs are already strongest.
Correlated RWA pairs already exist
Correlated onchain liquidity began with crypto native assets. But the first correlated markets for tokenized equities exist today: ten tokenized stocks trading against SPY, in Uniswap pools on Robinhood Chain.
In their first twelve days, these pools did $33 million in volume from more than 11,000 traders, much of that while US markets were closed. Some trades went straight from one stock to another, never touching dollars.

It’s worth mentioning we’ve also started seeing memecoins paired against “correlated” stocks: Elon memes paired against Tesla stock, hotdog memes against Costco stock. Unclear how correlated these will actually be in terms of price, but I guess “vibes” is another type of correlation.
AMMs will win
Correlated pairs are just one part of the puzzle. The other part is AMM design. Uniswap v4 hooks enable full market customizations that can significantly improve LP returns. Our DualPool hook, for example, puts a pool's idle liquidity to work earning lending yield between swaps.
Despite ~$4.6T in volume on Uniswap, I believe AMMs are still in their infancy. There are many promising paths that will improve their competitiveness, being built by Uniswap Labs, our partners, and the ecosystem. A lot more here coming soon.
The case against index funds in 1976 was that a fund making no decisions could never beat professionals paid to make them. Fifty years later, the fund that makes no decisions beats about 90% of the professionals. More importantly, index funds democratized investing and improved the lives of everyday people. I believe passive liquidity will win with a similar playbook, and have an even greater impact, by dramatically lowering barriers to creating and participating in markets.

Correlated Pairs: How AMMs Win the Biggest Markets

Written by @haydenzadams, Founder of @Uniswap.

I’ve spent 9 years working on the frontier of DeFi. It’s a fascinating surface area with infinite depth and the capability to transform capital markets.
I’ve always believed in the massive potential of the AMM, but for the past decade, one big question nagged at me. Can this new market structure really become the core engine for all financial markets?
It’s taken years of evolution and growth, but there’s an increasingly clear path to AMM global dominance forming. The best way to explain it starts in 1976.
Tokenization changes who makes markets
The index fund turns 50 this month. When Jack Bogle launched it in 1976, he hoped to raise $150 million. He collected $11.3 million. Rivals called it "Bogle's folly" and printed posters calling index funds un-American. A fund that made no decisions, they argued, could never beat professionals paid to make them. Today the majority of American fund assets sit in passive vehicles.
I've been thinking about this lately, because tokenization’s folly moment is ending. The SEC has approved Nasdaq and NYSE to trade tokenized shares. DTCC, which settles nearly every security in America, ran a live trial of tokenized trading in July. Nearly all this activity is described the same way: tokenization as an infrastructure upgrade.
Same markets, faster, cheaper, always on. This is all true, but I think the upgrade framing buries the bigger story. Tokenization makes markets programmable: changing what markets exist, who makes them, and what those markets trade against.
In 2018 I built Uniswap, a protocol that automates market making. Anyone can deposit two assets into a shared pool and earn a fee on every trade, while prices adjust along a curve as people buy and sell. Uniswap has run autonomously since the day it launched, settled more than $4.6 trillion in volume, and helped push decentralized exchanges from under 1% of centralized spot volume to over 20%.
As AMMs like Uniswap have grown, their liquidity has organized into a pattern most of finance hasn’t noticed yet: correlated pairs.
The easiest wins
To win at everything, you have to start by winning something. AMMs found product-market fit in the long tail, where most assets couldn't get a professional market maker's attention at all. On Uniswap, anyone could create a market in one transaction, with issuers and early supporters as the first LPs.
Stable pairs came next. On a pair like USDC/USDT, a good passive strategy gets close enough to optimal that a lower cost of capital more than covers the gap. This is why professional trading firms don’t bother market making these stable swaps today. They’re being undercut by passive AMMs.
Big margins, no competition
Traditional financial markets fully belong to market-making firms. They bundle capital, trading strategy, execution technology, settlement, and distribution into one vertically integrated business. The architecture evolved for good reasons. Assets lived in separate systems, settlement was slow, and someone had to perform every function, so one firm performed them all.
With enough scale, all those fixed costs pay for themselves. Citadel Securities handles roughly 25% of American equity volume and produced a record $12.2 billion in net trading revenue last year on roughly $21 billion of trading capital.
Most people read those numbers as proof the system works. I read them as entrenchment.
Breaking the bundle
Blockchains create competition at each layer, breaking the bundle apart. Execution happens through code. Custody and settlement are shared services anyone can plug into. What once required proprietary infrastructure is now open source software.
With AMMs, capital is the scarce input and the edge belongs to whoever holds inventory most cheaply. A trading firm needs high returns to justify its overhead, so an LP willing to accept less undercuts it. Most market makers hedge away all price exposure, and hedging costs money, so an investor who already holds the assets takes that exposure for free. And asset issuers have a negative cost of capital, since asset issuers typically have to pay professional market makers to market make on their new assets.
Put simply, DeFi and AMMs lower the barrier to making markets, opening the space up to many new participants. Their edge can come from many different sources such as a lower cost of capital, desiring the inventory exposure professional firms would typically hedge, or even being the issuer themselves.
But everything rests on one question: can automated strategies perform well enough for this to hold?
Liquidity follows correlation
Recently, I was on a call with one of the largest institutions in finance. They asked me what base pairs were most common in DeFi. I explained that Ethereum-based assets tend to trade against ETH, Solana assets against SOL, and stablecoins pair against each other, with a small number of highly liquid pairs bridging between these clusters.

No one designed that. It emerged organically, in part because LPs do best when the assets they hold move together. Correlation means less inventory risk for liquidity providers, deepening liquidity. As assets tokenize, the biggest markets in the world will reorganize the same way.
They can't today. Traditional markets settle overwhelmingly in dollars out of necessity. Assets live in siloed systems, and fiat rails like SWIFT and Fedwire are the glue that holds everything together. But blockchains are a far more expressive glue. Tokenize the assets, and they share a settlement layer, so any asset can trade directly against any other.
NVDA/USD can become NVDA/SPY, with SPY/USD as the bridge back to dollars. Oil companies can trade against an oil ETF or tokenized oil. Private credit can trade against tokenized Treasury funds. Tokenization also enables markets that span different types of assets, which is extremely difficult, if not infeasible, with TradFi infrastructure.
Delta neutrality is inefficiency
Traditional market-making firms generally try to be “delta neutral,” which is fancy traderspeak for denominating in dollars and wanting to minimize any non-dollar risk. When making a volatile asset they'll pay money to reduce their non-dollar risk (aka hedging), usually through options. This is one of the higher cost aspects of traditional market making.
Pairing assets into lower volatility “correlated pairs” connected by a few higher volatility “bridge pairs” brings numerous efficiency unlocks, but the most important one is that market making is cheaper and more efficient if the people market making on the assets actually want to hold the underlying assets.
And the more correlated a pair, the smaller any gap between today’s passive AMM strategies and the most sophisticated active strategies, making it easier to “undercut” them with this lower inventory cost.
To make it concrete, if you're long NVIDIA, you’re probably also long SPY, and the gap in efficiency between passive AMMs and active strategies is much lower for NVIDIA/SPY than for NVIDA/USD.
Connected liquidity
If stocks trade against SPY instead, every trade that starts or ends in dollars routes through the same pair: SPY/USD. These bridge pairs still require sophistication. However, there are much fewer of them, and they carry so much flow that professional attention is worth it.
DeFi already runs this way. ETH/USDC is one of the deepest markets onchain because every cluster routes through it. Passive LPs supply the correlated pairs, while active LPs compete over the bridge pairs.
Investors can still buy and sell everything in dollars, since routing across pools is automatic. And liquidity will concentrate where risk is lowest, not where legacy plumbing requires it to sit. That pushes the deepest markets into correlated pairs, the ground where AMMs are already strongest.
Correlated RWA pairs already exist
Correlated onchain liquidity began with crypto native assets. But the first correlated markets for tokenized equities exist today: ten tokenized stocks trading against SPY, in Uniswap pools on Robinhood Chain.
In their first twelve days, these pools did $33 million in volume from more than 11,000 traders, much of that while US markets were closed. Some trades went straight from one stock to another, never touching dollars.

It’s worth mentioning we’ve also started seeing memecoins paired against “correlated” stocks: Elon memes paired against Tesla stock, hotdog memes against Costco stock. Unclear how correlated these will actually be in terms of price, but I guess “vibes” is another type of correlation.
AMMs will win
Correlated pairs are just one part of the puzzle. The other part is AMM design. Uniswap v4 hooks enable full market customizations that can significantly improve LP returns. Our DualPool hook, for example, puts a pool's idle liquidity to work earning lending yield between swaps.
Despite ~$4.6T in volume on Uniswap, I believe AMMs are still in their infancy. There are many promising paths that will improve their competitiveness, being built by Uniswap Labs, our partners, and the ecosystem. A lot more here coming soon.
The case against index funds in 1976 was that a fund making no decisions could never beat professionals paid to make them. Fifty years later, the fund that makes no decisions beats about 90% of the professionals. More importantly, index funds democratized investing and improved the lives of everyday people. I believe passive liquidity will win with a similar playbook, and have an even greater impact, by dramatically lowering barriers to creating and participating in markets.

Source:https://twitter.com/uniswap/status/2092384590325522630