Introduction
Automated Market Makers (AMMs) have transformed decentralized cryptocurrency trading by allowing users to exchange tokens directly against smart contracts rather than relying exclusively on traditional order books. However, not all automated market makers distribute liquidity in the same way.
The traditional Uniswap V2 model distributes liquidity across the entire theoretical price curve, from zero to infinity. This design is simple, robust, and relatively passive: liquidity providers do not need to predict a particular trading range to keep their capital available.
Newer models, particularly Uniswap V3 and Uniswap V4, introduce concentrated liquidity. Instead of distributing capital across every possible price, liquidity providers can select a specific interval in which their capital will be available for trading.
This innovation can dramatically increase capital efficiency. Yet it also introduces a fundamental trade-off: the narrower the selected price range, the more dependent the position becomes on price stability, trading volume, and active management.
Extremely narrow liquidity positions take this concept to its logical extreme. They can offer substantial fee-generating potential per unit of capital while the market trades inside the selected range. Conversely, a relatively small price movement can push the position completely out of range, leaving it holding only one asset and earning no further trading fees until the price returns.
Understanding this difference requires examining not only the practical behavior of these pools but also the mathematical models governing their liquidity, pricing, capital requirements, and returns.
1. The Traditional AMM Model: Uniswap V2
1.1 The constant-product equation
The mathematical foundation of a traditional Uniswap V2 pool is the constant-product invariant:
x × y = k
Where:
xis the reserve quantity of Token X.yis the reserve quantity of Token Y.kis the constant product of the reserves.
For example, imagine a pool containing:
- 10 BNB
- 6,000 USDT
The initial reserve ratio implies a spot price of:
P = y / x = 6,000 / 10 = 600
Therefore, the pool’s initial marginal price is 600 USDT per BNB.
The constant product is:
k = 10 × 6,000 = 60,000
When a trader buys BNB with USDT, the pool receives USDT and releases BNB. The reserves change, but the swap must satisfy the constant-product invariant, with fees handled according to the protocol’s swap rules.
For a trade involving an input amount Δy, ignoring fees for simplicity, the new BNB reserve is:
x' = k / (y + Δy)
The quantity of BNB received by the trader is:
Δx = x − x'
This nonlinear relationship explains why larger trades produce greater price impact relative to the pool’s available liquidity.
The more a trade changes the reserve ratio, the further the execution price moves from the initial spot price.
1.2 Why V2 distributes liquidity across the entire price curve
In the traditional constant-product model, the pool supports trading throughout the theoretical price interval:
0 < P < infinity
The pool does not require a liquidity provider to choose a lower or upper price boundary. Its reserves automatically adjust as traders exchange one asset for the other.
This design offers several advantages:
- Liquidity remains available as the market price changes.
- The pool does not require range selection or routine rebalancing.
- Multiple liquidity providers can participate in a shared reserve structure.
- The design is relatively straightforward to understand and maintain.
However, there is an important economic limitation: a significant portion of the capital may support price levels that rarely, or perhaps never, occur.
Consider a stablecoin pool whose market price normally fluctuates around $1.00. Liquidity allocated to extremely distant prices contributes to the pool’s overall curve even though most trading occurs near parity.
This means that capital is not used equally efficiently at every price.
The limitation becomes especially relevant when comparing a full-range pool with a concentrated pool that allocates much more liquidity around the prevailing market price.
1.3 How much capital does V2 use near the current price?
Let P denote the price of Token X in units of Token Y. In a fee-free constant-product model:
x = √(k / P)
y = √(k × P)
The total reserve value, measured in Token Y, is:
V(P) = P × x + y = 2 × √(k × P)
This equation assumes that the pool is valued at its current marginal price.
If the price moves from P0 to P1, the change in reserve quantities is determined by the square-root relationship rather than by a linear relationship.
The reserve composition is therefore continuously changing, even though the underlying invariant remains constant.
An important distinction is that V2’s liquidity is not literally divided into separate, independent buckets of capital for each price. It follows a single reserve curve. Describing the capital as being spread across all prices is a useful economic interpretation of that curve, not a statement that the same funds are independently available at every price simultaneously.
2. Concentrated Liquidity: The Fundamental Change in V3
Uniswap V3 introduced concentrated liquidity, allowing each liquidity provider to define a lower price boundary and an upper price boundary.
Instead of providing liquidity over the entire theoretical price range, a position operates within:
Pa ≤ P ≤ Pb
Where:
Pais the lower price boundary.Pbis the upper price boundary.Pis the current pool price.
The position’s liquidity is active only while the market price remains within its selected range.
The central innovation is that capital can be concentrated where trading is expected to occur.
A liquidity provider who believes BNB will trade between $590 and $610, for example, can allocate capital specifically to that interval rather than providing full-range liquidity.
This makes the position more capital-efficient within the selected range, but less resilient to price movements beyond its boundaries.
2.1 Virtual reserves and concentrated liquidity
Concentrated liquidity does not simply divide the pool into independent V2 pools with isolated reserves. Instead, each position behaves like a translated constant-product curve within its permitted price interval.
Define:
L = √(k_virtual)
Where L represents the position’s liquidity parameter and k_virtual is the constant product associated with its virtual reserves.
The virtual reserve quantities are:
xv = L / √P
yv = L × √P
Therefore:
xv × yv = L²
The key difference is that virtual reserves are mathematical quantities used to determine price behavior. They can be much larger than the actual tokens deposited by the liquidity provider.
The real token balances needed to support the position depend on its price boundaries.
For a position with lower price Pa, upper price Pb, and current price P, the real reserves while the position is in range are:
x = L × (1 / √P − 1 / √Pb)
y = L × (√P − √Pa)
These equations assume that Token X is the asset whose quantity decreases as its price rises, and that prices are consistently expressed in Token Y per Token X.
The formulas show why concentrated liquidity can support much larger effective reserves than the actual capital deposited.
A position only needs to hold enough Token X to support price movement toward its upper boundary, and enough Token Y to support movement toward its lower boundary.
Once either boundary is reached, the position’s composition becomes entirely one asset.
The mathematical basis for this model is described in the Uniswap V3 whitepaper and official developer documentation.
2.2 What happens outside the selected range?
A concentrated position has three possible states.
State A: Below the lower boundary
If P < Pa, the position consists entirely of Token X under the price convention used above. It is inactive and does not earn swap fees.
State B: Within the range
If Pa ≤ P ≤ Pb, the position contains both assets, its liquidity is active, and it can earn a share of trading fees.
State C: Above the upper boundary
If P > Pb, the position consists entirely of Token Y. It is inactive and does not earn swap fees.
The position is not automatically liquidated when the price leaves the range. Its assets remain in the position, but its liquidity stops participating in trades until the price returns to the selected interval.
This is one of the most important practical differences between V2 and V3.
A full-range V2 position remains available across the entire price curve. A narrow V3 position can become inactive after a relatively small market movement.
3. Capital Efficiency: Why Narrow Ranges Are Attractive
The main economic argument for concentrated liquidity is that it can produce more active liquidity from the same amount of capital.
Consider a simplified stablecoin example.
A liquidity provider has $10,000 available for a pair trading close to $1.00.
The provider could choose either:
- A full-range position.
- A narrow position between $0.99 and $1.01.
The narrow position concentrates its capital around the expected trading price.
Uniswap’s documentation gives an illustrative comparison in which a V2 stablecoin pool uses only around 0.50% of its total capital for trading between $0.99 and $1.01. This example illustrates the potential efficiency gain from concentrating liquidity; it should not be treated as a universal ratio for all pools.
The general principle can be represented as:
E = Effective liquidity in range / Capital deposited
Here, E is a conceptual capital-efficiency measure. A narrow position can have a much higher value of E near its current price than a full-range position with the same capital.
However, capital efficiency is not equivalent to investment profitability.
Higher active liquidity can improve execution for traders and increase the liquidity provider’s share of trading fees. But the position’s profitability also depends on:
- Trading volume while the position is active.
- The liquidity supplied by competing providers.
- The pool’s fee tier.
- The direction and magnitude of price movements.
- Rebalancing costs and gas fees.
- The value of the assets when the position is eventually withdrawn.
A highly efficient position that remains out of range for most of the month may generate less income than a less efficient position that remains active continuously.
4. The Mathematics of Narrow Price Ranges
To understand why extremely narrow positions behave differently, consider a pool whose current price is P0.
Suppose the provider selects a symmetric percentage range:
Pa = P0 × (1 − δ)
Pb = P0 × (1 + δ)
Where δ represents the relative distance from the current price to each boundary.
For example:
δ = 0.10: approximately ±10%.δ = 0.01: approximately ±1%.δ = 0.001: approximately ±0.1%.δ = 0.0001: approximately ±0.01%.
These examples describe symmetric percentage ranges, not necessarily the exact tick intervals available in a particular protocol.
The narrower the range, the less price movement the position can withstand before becoming inactive.
4.1 Approximate capital-efficiency scaling
For a range centered near the current price, capital efficiency generally increases approximately inversely with the range width when comparing positions under comparable conditions.
A useful approximation for a symmetric, narrow range with total width w, expressed as a fraction of the current price, is:
E ∝ 1 / w
For example, reducing the total range width from 2% to 0.2% can increase the theoretical capital efficiency by roughly an order of magnitude near the current price, under otherwise comparable assumptions.
This is a local approximation, not a guarantee that fee income will increase tenfold. Actual results depend on the precise liquidity distribution, price trajectory, and competing liquidity.
The economic benefit is concentrated in a smaller price interval. Outside that interval, the position’s effective liquidity is zero.
4.2 Price range and the probability of remaining active
Suppose the price fluctuates around a reference level and follows a simplified stochastic process. The probability that it remains within a given range decreases as the range narrows and the observation period increases.
Under a simplified Brownian-motion model, assume the logarithmic price follows:
d(ln P) = σ × dW
Where:
σis the volatility parameter.Wis standard Brownian motion.
For a symmetric logarithmic boundary with half-width a, the probability that the price has not crossed either boundary by time t, assuming the process starts at the center and has no drift, is determined by the following series:
S(t) = (4 / π) × Σ[n=0 to infinity] { [(-1)^n / (2n + 1)] × exp(−((2n + 1)^2 × π^2 × σ^2 × t) / (8 × a^2)) }
Here, S(t) represents the probability of remaining inside the interval throughout the entire period.
For a narrow percentage range, a is approximately the half-width in logarithmic price terms.
This model is deliberately simplified. Cryptocurrency prices exhibit jumps, changing volatility, trends, and other behaviors that violate the assumptions of standard Brownian motion.
Nevertheless, it reveals a crucial relationship: the probability of remaining in range depends strongly on the ratio between range width and volatility over time.
A range that appears safe during a quiet market can become extremely fragile during a volatile session.
The narrower the range, the more frequently the position is likely to require monitoring or adjustment.
5. Extremely Narrow Liquidity: When a Pool Starts Behaving Like a Trading Strategy
At very narrow ranges, concentrated liquidity begins to resemble an active market-making strategy rather than a passive investment.
The provider is effectively making a decision about where to buy one asset with the other and where to sell it back.
Consider a position centered on a stablecoin pair trading near $1.00.
A provider might choose a range between $0.9999 and $1.0001, corresponding to a total width of approximately 0.02% of the reference price.
If the market price stays inside this interval, the position can offer substantial liquidity relative to its deposited capital.
However, a price movement of only 0.01% from the center reaches one boundary. A movement beyond the boundary leaves the position entirely in one asset.
The position may then stop generating fees precisely when market activity becomes more directional.
This creates an important distinction between two concepts:
- Liquidity efficiency: How much active trading liquidity a position supplies per unit of capital.
- Liquidity availability over time: How consistently the position remains available to traders.
A narrow position may score highly on the first measure and poorly on the second.
5.1 Range orders and automatic asset conversion
A very narrow concentrated position can also resemble a limit order.
Suppose a provider supplies only Token X to a position above the current price. If the market rises through the position’s entire interval, the position can convert its Token X exposure into Token Y.
Similarly, a position below the market can convert Token Y into Token X as the market falls through the interval.
Unlike a traditional limit order, a concentrated position may earn trading fees while its liquidity is active. But there are important differences: execution occurs through the AMM’s continuous pricing mechanism, and the provider must manage the position if the price subsequently reverses.
If the position remains in place after a complete crossing, a reversal can effectively trade the assets back in the opposite direction.
Consequently, a narrow range can function as an automated conversion strategy, but it is not a guaranteed substitute for a conventional limit order.
6. Fee Income: Why Capital Efficiency Does Not Guarantee Higher Returns
Liquidity providers generally earn a share of the fees paid by traders who swap against their active liquidity.
A simplified model of fee income is:
F ≈ V × f × s
Where:
Fis the fee income over a period.Vis the trading volume executed against the pool.fis the swap fee rate.sis the provider’s effective share of active liquidity.
For example, suppose:
- The pool processes $1,000,000 in trading volume.
- The fee rate is 0.05%.
- The provider captures 10% of the applicable fee-generating liquidity.
Then the pool’s gross fees from that volume are approximately:
$1,000,000 × 0.0005 = $500
The provider’s illustrative share is:
$500 × 0.10 = $50
This calculation assumes that the specified volume is charged at the stated fee rate and that the provider’s 10% share remains constant throughout the period. It excludes protocol-level fee deductions, if any, and other costs.
In reality, the share s changes as liquidity providers enter or leave the active range and as the market price moves through different liquidity intervals.
A more realistic expression is:
F = ∫[0,T] f × s(t) × dV(t)
Where:
dV(t)is the incremental fee-generating volume.s(t)is the provider’s share of active liquidity at timet.Tis the measurement period.
If a position leaves its range, its active share becomes zero until it returns to the range or the provider adjusts the position.
Therefore, a narrow position’s realized fee income depends on how much trading volume occurs while it is active, not merely on how much capital efficiency it offers.
6.1 The competition effect
Suppose a provider concentrates liquidity around a price that appears particularly attractive.
Other providers may make the same decision.
As more liquidity accumulates in that interval, each provider’s proportional share of active liquidity decreases.
This can erode the expected advantage of the narrow range.
In a competitive pool, high apparent fee returns attract more capital. That additional capital increases available liquidity and divides the fee income among more participants.
The result is a dynamic equilibrium in which the most attractive ranges may also become the most competitive.
6.2 The cost of keeping liquidity active
To maintain a narrow position, providers may need to:
- Withdraw liquidity after the price exits the range.
- Create a new position around the updated market price.
- Exchange one asset for the other.
- Pay transaction and network fees.
- Account for swap fees and slippage when rebalancing.
If the position is managed on a network with inexpensive transactions, frequent adjustments may be feasible. On a network with expensive gas, the same strategy may become uneconomical.
A useful net-return model is:
Rnet = (F − Crebalance − Cgas − Cother) / V0
Where V0 is the initial capital value and the cost terms represent the relevant expenses over the period.
This is a simplified accounting expression, not a complete measure of investment performance. A full evaluation must also account for the changing value of the assets held in the position.
7. Impermanent Loss and the Cost of Providing Liquidity
Impermanent loss, also called divergence loss, describes how the value of an AMM liquidity position differs from the value of holding the same initial assets outside the pool.
It is not necessarily a realized loss in the accounting sense, and accumulated trading fees may offset it. However, the economic cost can remain significant.
7.1 The traditional V2 benchmark
For a standard constant-product position, assume the initial portfolio contains equal values of two assets and the price changes by a factor:
r = P1 / P0
Ignoring fees, the ratio of the final pool value to the value of holding the initial assets is:
VLP / Vhold = 2 × √r / (1 + r)
Thus, the relative divergence loss is:
IL = [2 × √r / (1 + r)] − 1
For example, if the price doubles, r = 2:
IL = (2 × √2 / 3) − 1 ≈ −5.72%
This means the pool position is worth approximately 5.72% less than simply holding the original asset quantities, before accounting for fees.
The comparison is relative to holding the same initial portfolio, not necessarily to the dollar value of the original deposit.
7.2 How concentrated liquidity changes the comparison
A concentrated position has a different inventory profile.
As the price rises toward the upper boundary, the position progressively sells Token X for Token Y. As the price falls toward the lower boundary, it progressively sells Token Y for Token X.
Once the price exits the range, the position consists entirely of one asset.
The economic consequences depend on the direction of the price movement, the position’s initial composition, and the chosen boundaries.
A concentrated position may generate more fees per unit of capital during favorable market conditions, but it can also accumulate a less desirable inventory when the price moves persistently in one direction.
For a price increase, a position may end up entirely in the asset that has appreciated less than the asset it sold along the way. For a price decrease, it may accumulate the depreciating asset.
The risk is particularly important for volatile token pairs.
7.3 Measuring the value of a concentrated position
Let the current price be P, with lower boundary Pa, upper boundary Pb, and liquidity parameter L.
Inside the range, the position value in Token Y is:
V(P) = P × x + y
Substituting the reserve equations gives:
V(P) = L × (2 × √P − P / √Pb − √Pa)
This expression applies while Pa ≤ P ≤ Pb, under the earlier token and price conventions.
Outside the range, the value must be calculated using the single remaining asset:
- Below the lower boundary, the position holds Token X.
- Above the upper boundary, the position holds Token Y.
This piecewise valuation is one reason concentrated-liquidity positions cannot be evaluated adequately using only the standard V2 impermanent-loss formula.
A narrow position’s value, fee income, and inventory risk must be modeled together.
8. V2 vs. V3: A Practical Comparison
| Feature | Traditional V2 | Concentrated liquidity |
|---|---|---|
| Price coverage | Full theoretical range | Selected price interval, or full range if configured |
| Capital efficiency near market price | Generally lower | Potentially much higher |
| Fee generation | Continues across price movements, subject to trading volume | Only while the position is active |
| Position boundaries | No custom boundaries | Provider-defined lower and upper prices |
| Asset composition | Changes continuously with the reserve ratio | Can become entirely one asset outside the range |
| Management | Usually more passive | May require monitoring and rebalancing |
| Exposure to price movement | AMM inventory and divergence risk | Inventory risk plus out-of-range inactivity |
| Suitable for narrow-range strategies | No custom range support | Yes |
| Complexity | Relatively simple | More sophisticated position management |
| Fee collection | Depends on the implementation | Often requires separate collection transactions in V3-style systems |
One important clarification: concentrated liquidity is not restricted to narrow ranges. A V3 or V4 provider can select a wide range or even a full-range position. The difference is that the provider has the option to choose where liquidity should be active.
9. An Illustrative Numerical Example
Consider two liquidity providers, each depositing $10,000 into the same token pair.
Provider A selects a full-range V2 pool.
Provider B selects a concentrated range around the current price.
Assume that both positions face the same trading market, but their active liquidity distributions differ.
For illustration, suppose that during a given period:
- The V2 position earns $80 in gross swap fees.
- The concentrated position earns $240 in gross swap fees.
- The concentrated position incurs $30 in rebalancing and transaction costs.
The resulting fee figures are:
| Metric | V2 position | Concentrated position |
|---|---|---|
| Initial capital | $10,000 | $10,000 |
| Gross fees | $80 | $240 |
| Rebalancing costs | $0 | $30 |
| Net fees before inventory changes | $80 | $210 |
| Net fee yield for the period | 0.8% | 2.1% |
These numbers are hypothetical and are not a forecast of actual pool performance.
The concentrated position generates more net fees in this example because it captures a greater share of trading volume while active.
But suppose the market then moves sharply outside the concentrated position’s range. Its fee income stops until the price returns or the position is adjusted.
The V2 position, by contrast, continues participating in trades across the price curve.
The example illustrates the central trade-off: concentrated liquidity can improve fee efficiency, but it exchanges passive price coverage for greater dependence on market behavior and management.
It is also important not to interpret the example as a direct comparison between two identical smart-contract implementations. Actual V2 and V3 pools can have different fee tiers, liquidity distributions, trading volumes, and arbitrage behavior. A rigorous comparison must control for those differences.
10. Why Extremely Narrow Ranges Can Be Fragile
Narrow ranges introduce several risks that are less pronounced in a traditional full-range pool.
10.1 Short periods of inactivity
A position may leave its range after a small price movement.
Even if the market later returns to its previous price, the position earns no fees while it is inactive.
For a position whose profitability depends on frequent small trades, these inactive periods can significantly reduce realized returns.
10.2 Sudden volatility
Cryptocurrency markets can experience rapid movements caused by liquidations, announcements, low liquidity, or large trades.
A position concentrated within a very small price interval may become inactive almost immediately during such an event.
The problem is not necessarily a smart-contract failure. It is a consequence of the strategy’s selected boundaries.
10.3 Rebalancing risk
Frequent rebalancing may appear to solve the inactivity problem, but it introduces a different set of risks.
The provider may repeatedly exchange assets after the price has already moved, paying transaction fees and accepting slippage.
If the market reverses shortly afterward, the provider may need to rebalance again.
A strategy that appears profitable when measured using historical volume may perform poorly once actual execution costs are included.
10.4 Toxic flow and informed trading
Liquidity providers face the possibility that traders or arbitrageurs act on information faster than the provider can adjust the position.
For example, if the external market price changes before the pool price is updated, arbitrageurs may trade against the stale pool price.
This can impose adverse selection costs on liquidity providers.
A narrow range does not eliminate this problem. In some circumstances, it can intensify the impact because a large proportion of the provider’s capital is exposed around a small interval.
10.5 Illiquid or volatile token pairs
For a pair involving a volatile or thinly traded token, a narrow price range can be especially difficult to maintain.
The market price may jump across several ticks, move rapidly in one direction, or experience large price changes from relatively small trades.
In such circumstances, high theoretical capital efficiency may be offset by low active time, substantial inventory changes, and expensive rebalancing.
11. Ticks and the Technical Limits of Narrow Ranges
Concentrated-liquidity protocols do not necessarily permit every arbitrary decimal price boundary.
In Uniswap V3, price boundaries are represented by discrete ticks.
The relationship between a tick i and the corresponding price is:
P(i) = 1.0001^i
This expression uses the protocol’s conventional raw token-price representation. The displayed price must account for token ordering and decimal normalization.
One tick corresponds to approximately a 0.01% multiplicative price change.
However, this does not mean every pool permits positions bounded by adjacent ticks. The minimum spacing between usable boundary ticks depends on the pool’s configured tick spacing, which is associated with the fee tier in Uniswap V3.
As a result, the narrowest available position depends on the particular pool and its configuration.
Even where a very narrow range is technically available, it may not be economically sensible. Tick boundaries are a protocol-level constraint; they do not guarantee that a position can remain active long enough to earn an adequate return.
12. Can Narrow-Range Liquidity Be Automated?
Yes. A liquidity-management bot can monitor the current price, determine whether a position is approaching a boundary, and execute a rebalancing strategy according to predefined rules.
A simple strategy might use the following steps:
- Read the current pool price and active tick.
- Calculate the position’s distance from its lower and upper boundaries.
- Estimate the potential fee income and cost of a new position.
- Withdraw or adjust liquidity when the expected benefit justifies the cost.
- Rebalance the token quantities if necessary.
- Create a new position around a target price or within a predetermined interval.
- Track realized fees, inventory value, transaction costs, and time spent out of range.
A more advanced strategy could incorporate volatility estimates, recent trading volume, liquidity distribution, and the probability of crossing the selected boundaries.
However, automation does not remove the underlying market risk.
A bot can react more quickly than a human operator, but it cannot guarantee that the next price movement will be favorable or that a position will earn enough fees to cover the costs of maintaining it.
Furthermore, a bot should not rely solely on the current pool price. It should account for external prices, transaction execution, slippage, and the possibility of rapid price changes between monitoring and transaction confirmation.
13. How to Evaluate a Concentrated-Liquidity Strategy Properly
A meaningful analysis should go beyond the advertised annual percentage yield (APY) or the apparent liquidity efficiency.
At a minimum, the following variables should be measured.
Active time
What percentage of the observation period did the position remain within its price range?
Fee income per unit of capital
How much fee income was generated relative to the actual capital deployed, rather than the much larger virtual liquidity value?
Rebalancing costs
How much was spent on gas, swaps, slippage, and position adjustments?
Inventory performance
How did the final portfolio value compare with simply holding the original assets?
Price volatility
How frequently did the market cross the selected boundaries?
Competition
How much liquidity was supplied by other providers within the same interval?
Execution quality
Were trades executed at favorable prices, or did arbitrage and adverse selection reduce the effective return?
Operational reliability
Did the management system correctly handle failed transactions, network congestion, stale prices, and changes in liquidity?
A complete profitability measure can be expressed as:
Π = F + VT − VH − C
Where:
Fis accumulated fee income.VTis the terminal value of the liquidity position’s assets, excluding fees if those are accounted for separately.VHis the terminal value of the initial portfolio had it simply been held.Crepresents transaction and other relevant costs.
The exact accounting must avoid double-counting fees or assets. The equation is intended to compare the strategy against a defined holding benchmark, not to imply that all returns can be captured by a single universal formula.
For practical testing, historical pool data should be replayed using realistic assumptions about execution costs, price movements, liquidity competition, and rebalancing delays.
A backtest that assumes perfect execution at every desired boundary will generally overstate achievable performance.
14. Which Model Is Better?
There is no universally superior model. The answer depends on the market, the investor’s objectives, and the ability to manage the position.
V2 may be preferable when:
- The priority is simplicity and relatively passive participation.
- The provider does not want to manage custom price boundaries.
- The market is volatile and the provider wants liquidity to remain available across a broad price range.
- The expected fee income does not justify frequent rebalancing.
- The provider values operational simplicity over capital efficiency.
Concentrated liquidity may be preferable when:
- Trading volume is concentrated within a reasonably predictable price interval.
- The provider can monitor and manage the position.
- The network’s transaction costs are low enough to support the strategy.
- The provider understands the consequences of becoming single-sided.
- Expected fee income compensates for inventory risk, inactivity, and management expenses.
Extremely narrow liquidity may be appropriate when:
- The pair has sufficiently stable prices or predictable short-term behavior.
- The range is selected using a defensible market model rather than an arbitrary percentage.
- Trading volume is high enough to justify the capital and operational costs.
- The provider can monitor range exits and execute adjustments reliably.
- Stress tests show that the strategy remains viable under realistic volatility and execution assumptions.
Stablecoin pairs are often natural candidates for narrower ranges because their prices tend to remain near parity. Nevertheless, stablecoins can lose their pegs, and even seemingly stable pairs can experience sudden volatility.
For volatile assets, a very narrow range may be more appropriately viewed as an active market-making strategy than as a passive yield-generating investment.
Conclusion
The evolution from Uniswap V2 to concentrated liquidity represents a fundamental change in how capital is allocated inside automated market makers.
The traditional constant-product model prioritizes broad price coverage and operational simplicity. Concentrated liquidity allows providers to allocate capital where they expect trading to occur, potentially improving execution quality and increasing fee income per unit of capital.
Mathematically, the innovation is elegant: a liquidity position behaves like a translated constant-product curve within a selected price interval, using virtual reserves to deliver substantial effective liquidity with a smaller amount of deposited capital.
Economically, however, the trade-off is unavoidable.
A narrower range can increase capital efficiency, but it also reduces the amount of price movement the position can tolerate before becoming inactive. Realized returns therefore depend not only on liquidity concentration but also on volatility, trading volume, competition, inventory changes, transaction costs, and management quality.
The most important distinction is this:
V2 prioritizes liquidity availability across the price curve. Concentrated liquidity prioritizes liquidity efficiency within a chosen price interval.
Neither design eliminates the fundamental risks of market making. Concentrated liquidity simply gives providers greater control over where those risks are taken.
For anyone evaluating narrow-range strategies, the relevant question is not merely how much liquidity can be created from a given amount of capital. It is how much net economic value that liquidity can generate after accounting for inactive periods, asset-price movements, fees, and the costs of keeping the position competitive.
That is the difference between theoretical capital efficiency and sustainable liquidity provision.
Further Reading
- Uniswap Developers — Concentrated Liquidity
https://developers.uniswap.org/docs/get-started/concepts/liquidity-providers/concentrated-liquidity - Uniswap Developers — Uniswap V2 Pricing
https://developers.uniswap.org/docs/protocols/v2/concepts/pricing - Uniswap V3 Core Whitepaper
https://app.uniswap.org/whitepaper-v3.pdf - Uniswap Blog — A Primer on Uniswap V3 Math, Part 2
https://blog.uniswap.org/uniswap-v3-math-primer-2 - Uniswap Developers — Understanding Returns
https://developers.uniswap.org/docs/protocols/v2/concepts/understanding-returns
Disclaimer: This article is for educational purposes only. It does not constitute financial or investment advice. Liquidity provision involves market risk, smart-contract risk, and the possibility of losses relative to holding the underlying assets.