Impermanent Loss Explained for Launch Teams (With the Math)

Impermanent loss is the gap between what your liquidity position is worth and what the same two assets would be worth if you had simply held them. It happens because a constant-product pool sells the asset that is going up and accumulates the one that is going down, so you always end up holding more of the loser and less of the winner. The loss is "impermanent" only while the price is away from your entry — withdraw at that moment and it becomes permanent.

For a launch team this is not an abstract DeFi concern. You are the first liquidity provider, you are providing the pair that everyone else trades against, and your entry price is the one you set.

The mechanics in one worked example

You seed a pool with 1,000 USDC and 1,000,000 tokens at an opening price of $0.001 per token. The constant-product invariant is:

k = 1000 × 1,000,000 = 1,000,000,000

The token doubles to $0.002. Arbitrageurs buy tokens out of the pool until its internal price matches the market. The new reserves are:

tokens  = sqrt(k / price) = sqrt(1e9 / 0.002) = 707,107
USDC    = sqrt(k × price) = sqrt(1e9 × 0.002) = 1,414.21

Your LP position is worth both sides, and at equilibrium they are equal in value:

	Value
LP position	2 × 1,414.21 = 2,828.43 USDC
Simply holding	1,000 USDC + 1,000,000 tokens × $0.002 = 3,000.00 USDC
Impermanent loss	−171.57 USDC, or −5.72%

You did not lose money in absolute terms — 2,828 > 1,000 — but you underperformed holding by 5.72% of the position's value. That distinction is where most of the confusion about this topic lives.

The formula, and what it says

For a price change ratio p (new price divided by entry price), the loss relative to holding is:

IL = (2 × sqrt(p)) / (1 + p) − 1

It is symmetric: a 2× move and a 0.5× move produce the same loss, because the pool's rebalancing is directionless.

Price change	Impermanent loss	On a $25,000 position
1.25×	−0.62%	−155 USDC
1.5×	−2.02%	−506 USDC
2×	−5.72%	−1,430 USDC
3×	−13.40%	−3,350 USDC
4×	−20.00%	−5,000 USDC
5×	−25.46%	−6,365 USDC
10×	−42.56%	−10,640 USDC

Two readings matter. First, the loss grows without bound as the price runs: a token that 100× costs an LP roughly −80%. Second, entering at 1.25× is nearly free — most of the damage comes from large moves, which is precisely the regime a new token lives in for its first week.

Why a new token is the worst case for a liquidity provider

An ETH/USDC pool on a mature chain moves a few percent a day. Your token might move 400% in an afternoon, and every part of that move is a trade against your position. Three compounding factors:

1. The pool is the only market. Early on, every buy takes tokens out of your reserves and every sell puts them back. You are the counterparty to all of it.
2. The price discovery is one-directional. A new token has no established fair value, so the first meaningful flow often runs far past any sensible level, then retraces. A round trip to 4× and back to 1× costs you nothing in IL terms — but the round trip through 4× with fees earned is generally worse than holding, because you sold tokens into the rise.
3. Fees are the only offset. The pool earns a fee on every swap, and that fee accrues to your LP tokens. On a high-volume new token the fee income can exceed the IL; on a quiet one it will not.

Net result = trading fees + any token incentives − impermanent loss. That is the only equation that matters, and every term in it is measurable except the future.

The reframing that helps launch teams

Here is the part most explainers skip. When you seed a pool with a token you minted, IL is an odd way to describe what is happening, because "holding" was never the alternative you were considering.

What the pool actually does is convert your token into USDC at market prices, continuously. When the price rises, the pool sells more of your token; when it falls, the pool buys some back. If you seeded 20% of supply and the pool is now 60% USDC, you did not "lose" 5.72% — you sold your way into a USDC position at prices you chose, with the price path decided by the market.

Seen that way, the real question is not "how do I avoid impermanent loss?" but:

• How much supply am I willing to sell, and at what prices? A pool puts the answer in the hands of buyers.
• Is the remaining supply going to be sold too? Dilution afterwards makes the pool's sales look cheap.
• Can I afford to have that USDC locked in the pool? It is your working capital until the lock expires.

This framing also explains why teams with a token they would rather not sell should not over-provision liquidity. A pool with 40% of supply in it is a machine for distributing 40% of supply.

How to reduce impermanent loss without pretending it disappears

You cannot eliminate IL in a constant-product pool. You can manage it:

Approach	Effect	Trade-off
Smaller position	Smaller absolute IL	Less depth, worse slippage for everyone
Deeper pool with real USDC	Same percentage IL, larger absolute numbers	More capital at risk
Choosing the entry price carefully	Decides what the pool sells at	Price is set by your liquidity ratio, not by hope
Reinvesting fees	Offsets IL directly	Fees are unpredictable for a new token
Concentrated liquidity (Uniswap V3)	Higher fee capture in a range	IL accelerates sharply when the price leaves the range
Holding the position instead	Removes IL entirely	No market for your token, or someone else's

That last row is the honest one. A token with no pool has no IL and also no price. If you are providing liquidity, you are exchanging a slice of your upside for a market — and if you do not want that trade, the answer is to seed less and accept a thinner market rather than to look for a trick.

A note on V3: on Arc Mainnet, Uniswap V3 and V4 are deployed alongside V2, and concentrated liquidity changes the IL profile materially — you can earn far more fees for the same depth, and you can also be fully converted to one side of the pair the moment the price exits your range. The Arctools liquidity tools target V2 pairs because the pair model is the easiest to verify and the most predictable for a first launch; see Uniswap on Arc Mainnet for how the versions compare.

A pre-launch IL sanity check

Before you commit capital, answer these five questions with numbers:

1. What is my position size in USDC terms? IL percentages apply to that number; the absolute figure is the only one that affects your treasury.
2. What is a plausible bullish move in the first month? A 4× costs 20% of the position.
3. What is a plausible bearish move? Halving costs 5.72% — the formula is symmetric, and the downside case is often forgotten because it feels less urgent.
4. What fee volume do I need to break even? On a typical 0.3% fee tier, you need roughly IL / 0.003 in swap volume to offset a given loss. A 20% IL on a 25,000 USDC position is a 5,000 USDC loss, which needs about 1.67M USDC of volume to cover.
5. Can I afford to have the capital illiquid? If you lock the LP tokens for twelve months, the answer has to be yes — locking and burning describe the commitment options.

Run the numbers before you sign, not after. The Arc Liquidity Pool Creator shows the opening price and depth you are about to create, and Add Liquidity reads live reserves so a subsequent addition matches the pool's current ratio instead of rebalancing you on entry.

The one-sentence version

Impermanent loss is the cost of being the counterparty to your own token's price discovery, it is larger for newer and more volatile assets, it is offset only by fees, and for a launch team the more useful question is how much supply you are willing to convert into USDC — which is a tokenomics decision dressed up as a maths problem.

To see the same arithmetic from the buyer's side of the pool, read Uniswap on Arc Mainnet. Arctools charges a flat 50 USDC to create or adjust a liquidity position and takes no share of your pool. Arctools is not affiliated with Circle or the Arc Foundation.