Impermanent Loss in Concentrated Liquidity: The Math, the Myths, and How to Model It
By RangeScout Research · 6 min read · 2026-03-24
Impermanent loss is worse in concentrated liquidity pools than in Uniswap V2 — but it is also fully modelable. Here is the exact formula, why "IL recovery" is mostly a myth, and how to stress-test a position before you open it.
The actual formula nobody shows you
For a full-range V2-style pool, impermanent loss at price ratio k (new price / old price) is:
IL(k) = 2·√k / (1 + k) − 1
At k=2 (price doubles), IL = −5.7%. At k=4, IL = −20%. At k=0.5, IL = −5.7% again — it's symmetric.
Concentrated liquidity makes this significantly worse. When you set a range [p_low, p_high] and price moves *outside* the range, your position converts entirely to one asset and stops earning fees. IL does not stop: the position is frozen while a buy-and-hold portfolio keeps moving with price, so the gap to holding keeps widening the further price travels. You're now holding a 100% directional bag with no upside from the pool.
Inside the range, the effective IL is amplified by your concentration factor, which for a tight range around the current price can be 10-50x. That's why a 2% price move in a concentrated position can produce more IL than a 20% move in a V2 pool. This applies equally on Uniswap V3, Meteora DLMM, Orca Whirlpools, and every other concentrated liquidity protocol.
The "IL recovery" myth
You'll see influencers claim "IL is temporary — if price comes back, your IL reverses." That's true for passive V2 positions, but it's misleading for concentrated liquidity for two reasons:
- You rebalance. Every rebalance locks in IL at the current price. If you rebalanced while ETH was at $2,400 and it's now back to $3,000, you realized the loss and can't recover it. Same story for SOL, ARB, or any volatile asset.
- Price paths matter. Two paths with the same endpoint produce wildly different IL. A smooth move to k=1.5 and back produces almost no IL. A volatile whipsaw that crosses your range three times and back produces cumulative IL that never reverses because each out-of-range event forced a rebalance.
The only honest way to model IL is to simulate thousands of realistic price paths — not assume price comes back.
Stress-testing before you deposit
RangeScout runs a scenario stress test on every range it recommends: what is your IL if price drops 20%? Jumps 50%? Has 3x its historical volatility for the next week? You see the numbers *before* you click deposit, not after you've lost money.
Try it on any pool across 10 supported chains — Uniswap V3, Meteora, Orca, PancakeSwap V3, Trader Joe, or any concentrated liquidity protocol. Paste an address here and get the full IL scenario grid in a few minutes.