You deposit $1,000 into an ETH/USDC pool. ETH doubles over the next month. Trading fees roll in. You withdraw feeling good — and discover you're worth less than if you'd simply held the same tokens in your wallet.

That gap is impermanent loss (IL). It's the most misunderstood number in DeFi liquidity provision, partly because the name sounds temporary and harmless, and partly because fee earnings can mask it until you actually compare outcomes.

This article explains IL without hand-waving: what it measures, when it bites, how Uniswap V2, V3, and V4 change the picture, and how to model scenarios before you commit capital. If you already know V3's range mechanics, our earlier piece on concentrated liquidity pairs well with this one.

What Impermanent Loss Actually Is

Impermanent loss is the difference between:

IL is usually expressed as a percentage. A −5% IL means your LP position is worth 5% less than a simple hold strategy at the same moment — before counting trading fees you've earned.

Important clarifications upfront:

A Concrete Example (V2 Math)

Suppose ETH is $2,000 when you deposit. You add $1,000 total: 0.25 ETH + 500 USDC.

ETH rises to $4,000 — a 2× move. The pool's constant-product formula (x × y = k) forces your position to hold less ETH and more USDC than when you started. After rebalancing:

You still made money in dollar terms — $1,414 beats your $1,000 deposit. But you underperformed a passive hold by about $86. That's IL.

The pattern generalizes: the further price drifts from your entry ratio, the larger IL grows. A 3× move costs more than a 2× move; a 5× move hurts more still. Symmetrically, IL exists when price falls — not only when it rises.

Why Pools Rebalance Against You

AMMs exist to quote prices continuously. When traders buy ETH from the pool, they're taking ETH out and leaving USDC in. Your share of the pool silently sells the appreciating asset and accumulates the other side.

That's the trade you make as an LP: you earn a cut of swap fees, and in exchange you let the pool auto-rebalance your stack. When one token outruns the other, rebalancing means you end up with less of the winner than a holder would keep.

Fees compensate you for providing that service. Whether compensation exceeds IL is the entire game.

When Fees Beat IL (and When They Don't)

Net LP return ≈ fee income − impermanent loss ± other incentives (e.g. token emissions).

Scenarios where LP often still wins:

Scenarios where IL dominates:

The mistake is entering a pool because the fee APR looks high without simulating IL for realistic price paths. APR billboards rarely show the hold comparison.

IL Across Uniswap Versions

Uniswap V2 — Full Range, Predictable Curve

V2 spreads liquidity from price zero to infinity. IL follows the classic constant-product curve and depends only on how far the price ratio moved since deposit. No range management — but also no capital efficiency upside.

For a given price change, V2 IL is well-studied and smooth. A 2× move ≈ −5.7%; a 3× move ≈ −13.4%; a 5× move ≈ −25.5%. These numbers assume no fees and no emissions — useful baselines, not net P&L.

Uniswap V3 — Concentrated Ranges Change Everything

V3 lets you concentrate capital between a min and max price. Inside the range, IL behaves like a magnified V2 position — narrower range, higher sensitivity to price moves. Outside the range, your position converts entirely to one token and stops earning swap fees until you rebalance.

Three V3-specific IL gotchas:

Read our dedicated walkthrough: Uniswap V3 Concentrated Liquidity Explained.

Uniswap V4 — Hooks Add Variability

V4's singleton architecture and hook system let pool creators customize fee logic, dynamic curves, and even automated rebalancing strategies. IL is no longer locked to a single x × y = k shape — it depends on the hook-defined behavior of each pool.

Practical takeaway: before LP'ing into a V4 pool, read what its hooks actually do. A pool that auto-compounds fees or adjusts ranges may reduce effective IL; one with exotic curve logic may introduce risks a V2 calculator won't capture.

IL vs. Other Risks (Don't Conflate Them)

Liquidity providers juggle several distinct risks. Keeping them separate helps you reason clearly:

IL gets the headlines because it's unique to AMM LP'ing. But a −3% IL on a token that itself dropped 60% is cold comfort — you're still down heavily in absolute terms.

How to Estimate IL Before You Deposit

Back-of-envelope checks that take five minutes:

  1. Pick two or three price scenarios — e.g. −30%, flat, +50%, +100% from today.
  2. Calculate hold value of your intended deposit at each scenario.
  3. Estimate LP value using a protocol-appropriate IL formula or calculator.
  4. Subtract IL from expected fee income over your planned holding period.
  5. Add gas for entry, exit, and any planned rebalances — especially on Ethereum mainnet.

If the fee estimate doesn't clear IL in your bearish scenario, you're betting on price stability or volume spikes — know that's what you're doing.

Using DeFi Seeker to Model Scenarios

We built DeFi Seeker because spreadsheet IL math gets tedious fast, especially once V3 price ranges and V4 hook behavior enter the picture. The app supports Uniswap V2, V3, and V4 calculations on iPhone, iPad, and Mac.

What it's useful for in practice:

It's a planning tool, not a guarantee. Markets move faster than any model — but entering a pool with no model at all is worse.

A Pre-Deposit Checklist

Before you add liquidity anywhere:

If you can't answer the first two, pause. IL isn't a reason to avoid LP'ing entirely — plenty of positions work — but it is a reason to stop treating "high APR" as free money.

Bottom Line

Impermanent loss is the price of automated market making: your stack gets rebalanced as traders move price, and that rebalancing lags a simple hold when one asset runs. The loss is real while prices stay apart, fees can offset it, and V3/V4 add range and hook complexity on top.

Model before you deposit. Compare hold vs. LP at the prices you actually expect — not just the optimistic ones. Tools like DeFi Seeker exist to make that comparison fast enough that you'll actually do it.