Skip to content

Liquidity Pools

Liquidity pools are the backbone of decentralized exchanges. Unlike Centralized Exchanges (CEX) where market makers place orders on a book, Decentralized Exchanges (DEX) use Smart Contracts holding paired assets that anyone can trade against.

No intermediaries. No permission required. No recourse if issues arise.

TL;DR

  • Liquidity pools replace order books—trade against a smart contract, not other traders
  • Constant product formula (x × y = k): larger trades = worse prices
  • Impermanent loss is the cost when prices diverge; arbitrageurs profit, LPs pay
  • Concentrated liquidity (V3/V4) is more capital efficient but amplifies IL risk
  • High APY is usually temporary token incentives, low TVL inflation, or IL-ignorant math
  • Before LP'ing: verify fees exceed IL, check LP token locks, stick to audited protocols

What Is a Liquidity Pool?

A liquidity pool is a smart contract holding two (or more) tokens that enables trading without a traditional order book. Instead of matching buyers with sellers, traders swap directly against the pool's reserves.

Figure 1: Basic liquidity pool mechanics.

graph LR
    subgraph "Liquidity Pool"
        POOL["Smart Contract<br/>━━━━━━━━━━━━<br/>ETH: 100<br/>USDC: 100,000"]
    end

    LP["Liquidity Provider"]
    TRADER["Trader"]

    LP -->|"Deposits ETH + USDC"| POOL
    POOL -->|"LP Tokens"| LP
    TRADER -->|"Swaps ETH for USDC"| POOL
    POOL -->|"Returns USDC"| TRADER

    style POOL fill:#E6E6FA
    style LP fill:#90EE90
    style TRADER fill:#FFD700

Key Components

COMPONENT FUNCTION
Smart Contract Holds tokens, executes swaps, enforces rules
Token Reserves The actual assets in the pool (e.g., ETH + USDC)
LP Tokens Receipt tokens proving your share of the pool
Pricing Formula Algorithm determining swap rates (usually x × y = k)

How It Differs from Order Books

On a CEX, you trade against other people's orders. On a DEX with liquidity pools, you trade against a pool of assets governed by math.

ASPECT ORDER BOOK (CEX) LIQUIDITY POOL (DEX)
Counterparty Other traders Smart contract
Price discovery Bid/ask matching Algorithmic formula
Liquidity source Market makers Anyone (LPs)
Withdrawal risk Exchange can freeze Permissionless
Execution Depends on orders Always available (if pool exists)

How Liquidity Pools Work

The Constant Product Formula

Most liquidity pools use the constant product formula, pioneered by Uniswap:

x × y = k

Where:

  • x: quantity of token A in the pool
  • y: quantity of token B in the pool
  • k: constant (remains unchanged through swaps)

Figure 2: Constant product formula in action.

graph TD
    subgraph "Initial State"
        I["ETH: 100 | USDC: 100,000<br/>k = 10,000,000<br/>Price: 1 ETH = 1,000 USDC"]
    end

    subgraph "After Swap"
        A["ETH: 110 | USDC: 90,909<br/>k = 10,000,000<br/>Price: 1 ETH = 826 USDC"]
    end

    SWAP["Trader sells 10 ETH<br/>Receives 9,091 USDC"]

    I --> SWAP --> A

    style I fill:#90EE90
    style SWAP fill:#FFD700
    style A fill:#FFB6C1

The formula enforces scarcity: as you buy one token, it becomes more expensive. As you sell, it becomes cheaper.

Providing Liquidity

To become a Liquidity Provider (LP), you deposit both tokens in the pool's current ratio. In return, you receive LP Tokens representing your share.

Figure 3: LP deposit and token minting flow.

sequenceDiagram
    participant You
    participant Pool
    participant LPToken as LP Token Contract

    You->>Pool: Deposit 1 ETH + 1,000 USDC
    Pool->>Pool: Verify correct ratio
    Pool->>LPToken: Mint LP tokens
    LPToken->>You: Receive LP tokens (1% of pool)
    Note over You: LP tokens prove ownership

When you withdraw, you burn your LP tokens and receive your proportional share of whatever's in the pool—which may be different from what you deposited.

Fee Distribution

Every swap pays a fee (typically 0.3% on Uniswap V2). This fee stays in the pool, increasing the value of LP tokens over time.

Figure 4: Fee distribution to LPs.

graph LR
    TRADER["Trader swaps<br/>1,000 USDC → ETH"]
    FEE["Fee: 3 USDC<br/>(0.3%)"]
    POOL["Pool reserves<br/>increase by 3 USDC"]
    LPS["All LP token holders<br/>share proportionally"]

    TRADER --> FEE --> POOL --> LPS

    style TRADER fill:#FFD700
    style FEE fill:#FFB6C1
    style POOL fill:#E6E6FA
    style LPS fill:#90EE90

Higher trading volume generates more fees. However, volume often correlates with volatility, which increases impermanent loss.

Impermanent Loss

Impermanent Loss (IL) is the cost of providing liquidity when token prices change. It's the difference between holding tokens in your wallet versus depositing them in a pool.

Not Actually "Impermanent"

The term is misleading. The loss becomes permanent upon withdrawal. It's only "impermanent" in that prices could theoretically return to their original ratio.

How It Happens

When prices diverge, arbitrageurs rebalance the pool. They profit from the price difference, and that profit comes from LPs.

Figure 5: Arbitrage-driven impermanent loss.

sequenceDiagram
    participant Market as External Market
    participant Arb as Arbitrageur
    participant Pool

    Note over Pool: ETH: 100 | USDC: 100,000<br/>Pool price: $1,000/ETH
    Note over Market: ETH price rises to $1,100

    Arb->>Pool: Buy cheap ETH from pool
    Pool->>Arb: Sells ETH at ~$1,000
    Arb->>Market: Sell ETH at $1,100
    Note over Arb: Profit: ~$100/ETH

    Note over Pool: ETH: 95.3 | USDC: 104,880<br/>Pool now matches market
    Note over Pool: LPs lost value to arbitrage

The Math

Impermanent loss depends on how much prices diverge from your entry point:

PRICE CHANGE IMPERMANENT LOSS
±25% 0.6%
±50% 2.0%
±100% (2x) 5.7%
±200% (3x) 13.4%
±400% (5x) 25.5%

Figure 6: Impermanent loss vs price divergence.

xychart-beta
    title "Impermanent Loss vs Price Change"
    x-axis "Price Ratio" [0.5, 0.75, 1.0, 1.5, 2.0, 3.0, 5.0]
    y-axis "IL (%)" 0 --> 30
    line [2.0, 0.6, 0, 2.0, 5.7, 13.4, 25.5]

A Concrete Example

You deposit 1 ETH ($1,000) + 1,000 USDC into a pool. Total value: $2,000.

Scenario: ETH doubles to $2,000

STRATEGY ETH USDC TOTAL VALUE
Just hold 1 ETH ($2,000) 1,000 USDC $3,000
LP position 0.707 ETH ($1,414) 1,414 USDC $2,828
Difference - - -$172 (5.7%)

You still made money ($2,828 vs $2,000 initial). But you made less than if you'd just held. That difference is impermanent loss.

When Impermanent Loss Hurts Most

Impermanent loss is most significant when:

  • One token appreciates significantly: You end up with less of the appreciating asset
  • One token depreciates significantly: You end up with more of the depreciating asset
  • High volatility: Constant rebalancing transfers value to arbitrageurs
  • Low fees: Trading fees don't compensate for IL

Impermanent loss is manageable when:

  • Prices stay stable: Stablecoin pairs (USDC/USDT) have minimal IL
  • High volume: Fees can exceed IL
  • Prices revert: If prices return to entry ratio, IL disappears

Major Protocols

Uniswap

The original Automated Market Maker (AMM). Launched in 2018, it introduced the constant product formula that most DEXs still use.

Version History:

VERSION LAUNCH KEY FEATURE
V1 2018 ETH pairs only
V2 2020 Any token pair, flash swaps
V3 2021 Concentrated liquidity
V4 2025 Hooks, singleton contract, native ETH

Uniswap V3: Concentrated Liquidity

V3 lets LPs specify price ranges for their liquidity instead of spreading it across all prices. This is known as concentrated liquidity.

Figure 7: V2 vs V3 liquidity distribution.

graph LR
    subgraph "V2: Full Range"
        V2["Liquidity spread $0 to ∞<br/>Capital efficiency: Low"]
    end

    subgraph "V3: Concentrated"
        V3["Liquidity in $900-$1,100 range<br/>Capital efficiency: High"]
    end

    V2 -->|"Evolution"| V3

    style V2 fill:#FFE4B5
    style V3 fill:#90EE90

Concentrated liquidity means more fees when price stays in range—but amplified impermanent loss if price moves outside your range.

Uniswap V4: Hooks

V4 introduces "hooks"—plugin contracts that execute before or after pool actions. Developers can customize:

  • Dynamic fees based on volatility
  • On-chain limit orders
  • MEV redistribution to LPs
  • Custom pricing curves

Curve Finance

Specializes in stablecoin and pegged asset swaps. Uses a modified formula (StableSwap) optimized for assets that should trade 1:1.

Figure 8: Curve vs Uniswap for stablecoins.

graph LR
    subgraph "Curve Advantage"
        STABLE["USDC/USDT/DAI pool"]
        LOW["Ultra-low slippage<br/>for stable pairs"]
    end

    subgraph "vs Uniswap"
        UNI["Same pair on Uniswap"]
        HIGH["Higher slippage<br/>constant product penalty"]
    end

    STABLE --> LOW
    UNI --> HIGH

    style LOW fill:#90EE90
    style HIGH fill:#FFB6C1

Curve pools have historically held billions in TVL because stable-to-stable swaps are a massive use case.

Curve Exploit (July 2023)

In July 2023, several Curve pools were exploited for approximately $70 million due to a Reentrancy vulnerability in the Vyper compiler. The exploit affected pools including alETH/ETH, CRV/ETH, and pETH/ETH. About $50 million was eventually recovered through white hat returns and negotiations.

Balancer

Allows pools with more than two tokens and custom weightings. A standard Uniswap pool is 50/50. Balancer pools can be 80/20, 60/20/20, or any configuration.

POOL TYPE USE CASE
50/50 Standard trading pairs
80/20 Reduce IL on volatile assets
Multi-token Index fund-like exposure

Raydium (Solana)

Combines AMM pools with an order book (Serum/OpenBook). Liquidity can be accessed both ways, theoretically providing better execution.

Liquidity Pool Risks

Smart Contract Risk

Pools are only as secure as their code. Exploits happen.

EXPLOIT TYPE MECHANISM
Reentrancy Contract called recursively before state updates
Oracle manipulation Price feeds exploited to drain pools
Flash Loan attacks Massive temporary capital used to manipulate prices
Admin key compromise Privileged functions exploited

Mitigation:

  • ☑ Stick to audited, battle-tested protocols
  • ☑ Check time since deployment (Lindy Effect)
  • ☑ Verify admin key setup (Multisig, Timelock)
  • ☒ Don't assume audited = safe

Rug Pulls

In permissionless systems, anyone can create a pool. Rug pull scam patterns:

  1. Deploy token with hidden minting function
  2. Create pool, provide initial liquidity
  3. Promote token, attract buyers
  4. Drain pool by minting tokens or removing liquidity

Figure 9: Anatomy of a rug pull.

sequenceDiagram
    participant Scammer
    participant Pool
    participant Victims

    Scammer->>Pool: Creates pool with SCAM/ETH
    Scammer->>Pool: Adds initial liquidity
    Note over Scammer: Promotes token

    Victims->>Pool: Buy SCAM token
    Note over Pool: Price rises

    Scammer->>Pool: Removes all liquidity
    Note over Pool: Pool emptied

    Victims->>Pool: Try to sell
    Note over Victims: No liquidity to exit

Red Flags:

  • ☒ Unlocked liquidity (LP tokens not locked or burned)
  • ☒ Anonymous team with no track record
  • ☒ Unverified contract source code
  • ☒ Admin functions that can mint, pause, or blacklist
  • ☒ Honeypot patterns (can buy but not sell)

Impermanent Loss (Revisited)

IL is inherent to the AMM model. LPs provide liquidity and pay the cost of IL in exchange for fees. The question is whether fees exceed IL.

For volatile pairs, they often don't.

MEV Extraction

Bots can front-run swaps, sandwich transactions, and extract value from pools. Maximal Extractable Value (MEV) is an inherent feature of public mempools.

Protection Options

Use private Mempools (Flashbots Protect), MEV-aware DEX aggregators (CoW Swap), or set tight slippage tolerances. None are perfect.

Evaluating Pools

Key Metrics

METRIC WHAT IT TELLS YOU
TVL Total Value Locked; larger = more stable pricing
Volume Trading activity; higher = more fee income
APR/APY Projected returns (often inflated by token incentives)
IL Exposure How volatile is the pair?
Protocol Age Lindy effect; older = battle-tested

APY Warnings

High APYs are almost always temporary or misleading:

  • Token incentives: Protocol pays you in their token, which often dumps
  • Low TVL: High fees on small pool = unsustainable
  • Calculation tricks: APY = (last 24h fees × 365), ignoring IL

Figure 10: High APY due diligence checklist.

graph TD
    HIGH["Pool showing 500% APY"]
    Q1["Is it from trading fees<br/>or token incentives?"]
    Q2["How long has APY<br/>been this high?"]
    Q3["What's the TVL?<br/>Is it sustainable?"]
    Q4["What's your IL exposure?"]

    HIGH --> Q1
    Q1 --> Q2
    Q2 --> Q3
    Q3 --> Q4

    style HIGH fill:#FF6B6B
    style Q1 fill:#FFE4B5
    style Q2 fill:#FFE4B5
    style Q3 fill:#FFE4B5
    style Q4 fill:#FFE4B5

Liquidity Lock Verification

Before buying tokens from a new pool, verify:

  1. Are LP tokens locked? (Check contract or locking service)
  2. For how long? (24 hours is meaningless; years matter)
  3. What percentage? (10% locked is still 90% rug risk)

Tools like DeFiLlama, DEXScreener, and TokenSniffer can help verify.

Liquidity Pool Strategies

Stable Pairs

Providing liquidity to stable-stable pairs (USDC/USDT, DAI/USDC):

  • ☑ Minimal impermanent loss
  • ☑ Consistent fee income
  • ☒ Lower APY than volatile pairs
  • ☒ Still exposed to smart contract and depeg risk

Correlated Assets

Pairs that move together (ETH/stETH, BTC/wBTC):

  • ☑ Reduced IL (prices should stay ~1:1)
  • ☑ Higher volume than stable pairs
  • △ Depeg events can cause significant IL

Concentrated Liquidity (V3)

Setting tight ranges on Uniswap V3:

  • ☑ Higher fee capture when in range
  • ☑ More capital efficient
  • ☒ Amplified IL if price exits range
  • ☒ Requires active management
  • ☒ Position earns nothing when out of range

Single-Sided Liquidity

Some protocols allow depositing one token:

  • ☑ Simpler UX
  • ☒ Protocol converts half to the other token anyway
  • ☒ You still have IL exposure

CEX Liquidity vs DEX Liquidity

Understanding where liquidity comes from matters for execution quality.

FACTOR CEX (MARKET MAKERS) DEX (LIQUIDITY POOLS)
Provider Professional firms Anyone with capital
Obligation None (in crypto) None (permissionless)
Withdrawal Can leave instantly Can withdraw anytime
Accountability Reputational Smart contract only
Crisis behavior Often withdraw Passive; always available
Large order execution ☑ Usually better ☒ Usually worse

During the October 2025 crash, CEX market makers withdrew liquidity within minutes. DEX pools remained—not because they're better, but because they're automated. No one can "withdraw" a smart contract.

This cuts both ways: passive availability during crashes, but no one managing risk or providing depth when you need it.

References

AMM Fundamentals

Impermanent Loss

Protocol Documentation

Security & Exploits

Concentrated Liquidity

Changelog

DATE AUTHOR NOTES
2025-12-24 Artificial. Generated by robots.
2026-01-04 Denizen. Reviewed, edited, and curated by humans.