🚀 What Makes a DEX More Capital Efficient?
Uniswap v3's concentrated liquidity claims up to 4,000x better capital efficiency than the old constant-product model. A recent Dune study covering H1 2026 shows the real-world picture is messier than that headline number suggests it should really be.
🔎 The Theory
Old-style AMMs — the model STONfi own pools still run on — spread liquidity across the entire price curve, zero to infinity. Concentrated liquidity lets LPs deposit only near the current price instead, so a narrow position can replace thousands of dollars of old-style liquidity for trades landing in range.
📊 What Actually Happens in Practice
Dune rebuilt roughly 6.5 million LP positions across Uniswap v3/v4, PancakeSwap v3, and Aerodrome Slipstream over 26 weekly snapshots. Finding: 29.5% of that capital — about $542M in a typical week — sat outside its active range earning nothing at all, a share that held steady all year. Over a third of that idle capital hadn't been adjusted in 90 days. Even Uniswap v4's hooks, built to redirect idle liquidity into yield, are used by only ~10% of v4's TVL, none of them producing yield from idle capital yet.
🧩 Where This Leaves a Constant-Product DEX Like STONfi
No out-of-range problem — capital simply can't sit idle outside a "range" that doesn't exist in this model at all. The trade-off runs the other way: no concentrated depth near the current price without deep enough TVL, and no active-management burden shifted onto LPs either.
🏁 Concentrated liquidity wins on paper. Constant-product avoids the exact failure mode the Dune data just exposed, at the cost of the upside.
$PUMP
Uniswap v3's concentrated liquidity claims up to 4,000x better capital efficiency than the old constant-product model. A recent Dune study covering H1 2026 shows the real-world picture is messier than that headline number suggests it should really be.
🔎 The Theory
Old-style AMMs — the model STONfi own pools still run on — spread liquidity across the entire price curve, zero to infinity. Concentrated liquidity lets LPs deposit only near the current price instead, so a narrow position can replace thousands of dollars of old-style liquidity for trades landing in range.
📊 What Actually Happens in Practice
Dune rebuilt roughly 6.5 million LP positions across Uniswap v3/v4, PancakeSwap v3, and Aerodrome Slipstream over 26 weekly snapshots. Finding: 29.5% of that capital — about $542M in a typical week — sat outside its active range earning nothing at all, a share that held steady all year. Over a third of that idle capital hadn't been adjusted in 90 days. Even Uniswap v4's hooks, built to redirect idle liquidity into yield, are used by only ~10% of v4's TVL, none of them producing yield from idle capital yet.
🧩 Where This Leaves a Constant-Product DEX Like STONfi
No out-of-range problem — capital simply can't sit idle outside a "range" that doesn't exist in this model at all. The trade-off runs the other way: no concentrated depth near the current price without deep enough TVL, and no active-management burden shifted onto LPs either.
🏁 Concentrated liquidity wins on paper. Constant-product avoids the exact failure mode the Dune data just exposed, at the cost of the upside.
$PUMP