Concentrated liquidity (V3) versus constant-product (V2)

The same capital, placed differently

A constant-product position quotes at every price the pair could ever reach. Most of that capital is reserved for prices that will not trade this year, which is why the depth a trader experiences near the current price is a small slice of the headline pool value.

Concentrated liquidity removes the assumption that a provider wants exposure everywhere. The v3 design gives providers the ability to bound their liquidity within an arbitrary price range, improving the pool's capital efficiency and letting them approximate a preferred reserves curve.¹ Same capital, narrower placement, more depth where trades happen.

Resist any single headline multiplier for that gain. It is entirely a function of how narrow the band is, so a number quoted without its range is not a fact about the model, it is a fact about somebody's chosen configuration.

The invariant, in one line

The mechanism is a translation rather than a new curve. A concentrated position's real reserves satisfy the relation that x plus L over the square root of the upper bound, multiplied by y plus L times the square root of the lower bound, equals L squared, with liquidity L equal to the square root of the constant-product k.¹ It is the familiar constant-product curve, shifted so that it terminates at the chosen bounds instead of running to the axes.

That is worth knowing because it explains the failure mode exactly. At the upper bound the position holds only the quote asset. At the lower bound it holds only the base asset. Those are not edge cases, they are the definition of the endpoints.

The position is an NFT, and that changes what you can do with it

Constant-product pool shares are fungible. Two providers in the same pool hold the same instrument, so shares can be pooled, staked, used as collateral or wrapped by anything that handles a standard fungible token.

A concentrated position is not fungible, because two positions in the same pool can have completely different bounds. The v3 design captures that by representing each position as an ERC-721 non-fungible token.¹ The standard itself is what makes each one uniquely identifiable and individually transferable.³

The practical consequence lands on composability. If your incentive programme, staking contract or treasury tooling expects a fungible balance, a concentrated position will not drop into it, and the integration work is real. Decide this before you choose the venue rather than after.

Out of range is out of the market

While the price sits inside the band, the position provides depth and earns fees. Once the price leaves, it provides neither. The capital is fully converted into whichever asset is on the wrong side and earns nothing until the price returns or somebody repositions.²

For a launch pool this is the decision that matters. A narrow band gives excellent depth for the price you expected and no depth at all for the price you got. Because the whole reason to hold depth on day one is the possibility of a move you did not expect, a band tight enough to look impressive in a model is often the wrong band to launch with.

Which one a launch pool should use

Ask two questions. Does somebody own the operational responsibility for monitoring and repositioning the band, with the authority and the tooling to do it at three in the morning. And does anything downstream need a fungible position token.

Two clear answers point to a concentrated position with a deliberately wide band around the opening price. Anything less than two clear answers points to constant product, accepting lower efficiency for a position that keeps quoting no matter where the price goes. In our experience the wrong choice here is almost never the curve; it is committing to a concentrated position that nobody was actually resourced to manage.