Impermanent Loss Calculator
Estimate the impermanent loss a constant-product AMM liquidity provider suffers when token prices diverge from the moment of deposit. Returns the percentage your LP position is worth less than simply holding the two tokens.
Last updated: May 2026
Compare with similar
About this calculator
Impermanent loss (IL) is the gap between the value of a liquidity provider's position in a constant-product automated market maker (e.g. Uniswap v2, SushiSwap) and the value of just holding the same two tokens outside the pool. It arises because arbitrageurs continuously rebalance the pool to track external prices, leaving LPs holding more of the falling token and less of the rising one. The standard formula for a 50/50 pool is IL% = (2 × √(rA × rB) − rA − rB) × 100, where rA = currentPriceA / initialPriceA and rB = currentPriceB / initialPriceB are the price ratios of each token since deposit. The result is always ≤ 0% — zero only when prices have not moved relative to each other. Key reference points: a 25% relative price change produces ~0.6% IL, a 100% change (2×) produces 5.7% IL, a 4× change produces 20% IL, and a 5× change produces 25.5% IL. Edge cases: the formula assumes a constant-product (x·y = k) pool with 50/50 weighting and no fees collected; with concentrated liquidity (Uniswap v3) IL can be dramatically larger because capital is leveraged within a tighter range. Trading-fee income is not subtracted here, so net P&L = fee yield − impermanent loss, and a profitable LP position requires fees to exceed IL over the holding period.
How to use
Example 1 — modest divergence. You deposit equal value of Token A at $100 and Token B at $1. After a month A rises to $150 (rA = 1.5) and B stays at $1 (rB = 1.0). Step 1: rA × rB = 1.5 × 1.0 = 1.5. Step 2: 2 × √1.5 = 2 × 1.2247 ≈ 2.4495. Step 3: rA + rB = 2.5. Step 4: (2.4495 − 2.5) × 100 ≈ −0.51%. Verify: a 50% one-sided move giving roughly 0.5% IL matches the published reference table — your LP is worth ~0.5% less than just holding the two tokens, which the pool's trading fees may or may not cover. ✓ Example 2 — severe divergence. You deposit ETH at $2,000 and a stablecoin USDC at $1. ETH rallies to $6,000 (rA = 3.0); USDC stays at $1 (rB = 1.0). Step 1: rA × rB = 3.0. Step 2: 2 × √3 = 2 × 1.7321 ≈ 3.4641. Step 3: rA + rB = 4.0. Step 4: (3.4641 − 4.0) × 100 ≈ −13.4%. Verify: a 3× one-sided move maps to ~13.4% IL in the standard IL table. On a $10,000 deposit, that's $1,340 less than simply holding — and unless the pool generated more than 13.4% in trading fees over the period, you would have been better off not providing liquidity. ✓
Frequently asked questions
Why is impermanent loss called 'impermanent' if I actually lose money?
The name comes from the fact that the loss is only realised when you withdraw from the pool — as long as you stay in, prices can revert and the paper loss disappears. If both tokens return to their original ratio, impermanent loss goes back to exactly zero regardless of how dramatic the round-trip price moves were. In practice this is misleading because prices rarely return to their starting ratio over a meaningful timeframe, and most LPs do withdraw at some point and lock in whatever loss exists at that moment. Many in the DeFi community now prefer the term 'divergence loss' because it is more accurate: the loss exists whenever prices diverge, and calling it impermanent encourages LPs to underestimate the risk. Treat any open IL as a real opportunity cost — every day you stay in the pool, you are betting that future fee revenue plus possible price convergence will outweigh the current paper deficit.
How does impermanent loss scale with the size of the price change?
IL is non-linear and grows much faster than the price change itself. A 1.25× price change produces only 0.6% loss, a 1.5× change produces 2.0%, a 2× change produces 5.7%, a 3× change produces 13.4%, a 4× change produces 20%, and a 5× change produces 25.5%. Importantly the formula is symmetric — a token going to 0.5× (down 50%) produces the same 5.7% loss as a token going to 2× (up 100%) relative to its pair. This is why providing liquidity for volatile-volatile pairs (like ETH/altcoin) carries much higher IL risk than stable-stable pairs (USDC/USDT), where prices rarely diverge more than a fraction of a percent. The convex shape also means that small price moves produce trivial IL, but large moves cause disproportionately large losses; this asymmetry is the core risk of constant-product liquidity provision.
When can trading fees offset impermanent loss?
Trading fees offset IL when the pool's annualised fee yield exceeds the IL incurred over the same period. Pools with high trading volume relative to total value locked (TVL) — typically major-pair pools on busy DEXes like ETH/USDC on Uniswap — can generate 10–50% APR in fees during active markets, which can comfortably exceed typical IL on moderate price moves. Pools with low volume relative to TVL, or pools holding tokens that experience large directional moves, are likely to leave LPs underwater. A useful rule of thumb: if you expect token prices to stay within ±25% of each other over your holding period, IL will be under 1% and any reasonable fee yield wins; if you expect one token to move 2× or more relative to the other, you will likely lose money LP-ing regardless of fees. Always compare fee APR against IL projections before depositing, and re-check periodically as price ratios change.
What are the common mistakes when evaluating impermanent loss?
The biggest mistake is treating IL as the only cost — you also pay gas fees to enter and exit the pool, suffer slippage on the deposit, and forgo any staking yield or appreciation you'd get from simply holding the tokens. People also forget that the standard IL formula applies only to constant-product 50/50 pools; concentrated-liquidity positions on Uniswap v3 can experience IL several times larger because liquidity is leveraged within a price range, and stepping out of that range turns your position into 100% of the underperforming token. Another error is comparing LP returns against the starting USD value of the deposit rather than against the value of just holding the two tokens — that comparison masks IL by counting price appreciation as 'LP profit'. People also assume past fee APR will continue, when fee yield is highly volatility- and volume-dependent and can collapse quickly. Finally, do not LP just because APR looks high; the highest-APR pools often pay it because they are the most likely to bleed LPs through IL or token depreciation.
When should I not use this impermanent loss calculator?
Do not use this calculator for concentrated-liquidity positions (Uniswap v3, PancakeSwap v3, Trader Joe Liquidity Book) — the standard constant-product IL formula does not apply and you need a v3-specific calculator that accounts for the chosen price range. It is not appropriate for weighted pools (Balancer 80/20 or 60/40), curve-stable pools that use a hybrid invariant, or single-sided staking arrangements, all of which have different IL profiles. Do not use it to estimate net LP profitability — you must separately model fee yield, gas costs, slippage, and any farming rewards or token emissions, then subtract IL from the gross. It is also not a real-time risk tool: it assumes specific entry and current prices and does not predict future IL, which depends on price paths you do not know in advance. For projecting expected IL over a holding period, you need scenario analysis or Monte Carlo simulation rather than a single-point calculator like this one.