MEV is profit extracted by reordering, including, or excluding transactions within blocks. It doubled from $550M in 2021 to $1.1B in 2024. For users, MEV acts as a hidden tax—worse execution prices, failed transactions, and value leakage to sophisticated bots.
What Is MEV?
Maximum Extractable Value (MEV) refers to the profit that block producers (validators, sequencers) and specialized searchers can extract by strategically ordering transactions. Because they control transaction order, they can profit at the expense of regular users.
Originally called "Miner Extractable Value" (when Ethereum used proof-of-work), the concept now applies to any blockchain where someone controls block construction.
Types of MEV
1. Sandwich Attacks (66% of MEV on Ethereum)
The most harmful form of MEV. Here's how it works:
- You submit a swap on Uniswap: buy ETH with 10,000 USDC
- A bot sees your pending transaction in the mempool
- The bot front-runs: buys ETH first, pushing the price up
- Your transaction executes at the worse (higher) price
- The bot back-runs: sells ETH at the new higher price
Result: You paid more than you should have. The bot pocketed the difference.
You set 10% slippage tolerance, expecting a max price of $1,980 per ETH. A sandwich bot front-runs you, pushing the price to $1,900. You buy at $1,900 instead of the expected $1,800—costing you an extra $100 that goes to the bot.
2. Arbitrage
Exploiting price differences across automated market makers (AMMs). Example: buy ETH at $2,500 on Uniswap, sell at $2,510 on SushiSwap, pocket $10 per ETH minus gas.
Unlike sandwich attacks, arbitrage is generally beneficial—it aligns prices across markets, improving overall market efficiency.
3. Liquidations
Racing to liquidate underwater lending positions. When a borrower's collateral falls below threshold, liquidators compete to close the position and claim the liquidation bonus (typically 5-15%).
Necessary for protocol health, but competition means most value flows to validators rather than distributed broadly.
MEV Supply Chain
The MEV ecosystem has specialized into distinct roles:
| Role | Function | Examples |
|---|---|---|
| Searchers | Find MEV opportunities, create bundles | Specialized trading firms, bots |
| Builders | Aggregate bundles, construct optimal blocks | Flashbots, BloXroute, Titan |
| Validators/Proposers | Select and propose the most profitable block | Lido, Coinbase, independent validators |
Over 90% of Ethereum validators now use MEV-Boost to outsource block building, participating in this supply chain.
Proposer-Builder Separation (PBS) Game Theory
Ethereum's transition to PBS — where block proposers (validators) outsource block construction to specialized builders — created a new economic game with important equilibrium properties.
The Builder Bidding Game
Builders compete in a sealed-bid auction each slot. The key dynamics:
- Information asymmetry: Builders who see more order flow (via private channels) can extract more MEV and bid higher
- Winner's curse: The winning builder bid is the highest, meaning they estimated the block's MEV value most aggressively — sometimes overbidding
- Revenue sharing: Competition between builders drives most MEV value to validators as builder profits get compressed
Two-Builder Duopoly
In practice, block building has concentrated into a small number of dominant builders. The theoretical model for a two-builder duopoly reveals concerning dynamics:
| Property | Competitive Outcome | Duopoly Reality |
|---|---|---|
| Builder profit margin | Approaches zero as competition increases | Small but positive due to exclusive order flow |
| Validator revenue | Captures ~99% of MEV value | Captures ~90-95% — builders retain edge from private order flow |
| Centralization risk | Minimal with many builders | Top 2-3 builders construct 90%+ of blocks — censorship and manipulation risks increase |
| Order flow incentive | Irrelevant with public mempool | Builders pay for exclusive order flow, creating centralization flywheel |
Builders with more exclusive order flow can extract more MEV, bid higher, win more blocks, and attract even more order flow — a centralizing feedback loop. This is why Flashbots proposed SUAVE (Single Unified Auction for Value Expression) and Ethereum researchers are exploring inclusion lists and other mechanisms to break builder dominance without sacrificing MEV protection.
JIT Liquidity as MEV Strategy
Just-In-Time (JIT) liquidity is a sophisticated MEV strategy where builders or searchers provide concentrated liquidity in the same block as a large swap, earning fees while removing liquidity before any adverse price movement:
- Searcher detects large pending swap in mempool
- Mints concentrated LP position around the swap's price range (front-run)
- Large swap executes, paying fees to the JIT position
- LP position is burned immediately after the swap (back-run)
JIT providers earn swap fees without bearing any inventory risk or impermanent loss — they exist only for the duration of a single block. This extracts value from passive LPs who provide liquidity across longer timeframes.
Solutions & Mitigation
Flashbots Protect
Routes your transactions directly to builders, bypassing the public mempool. Bots can't see your transaction, so they can't sandwich you. [Flashbots Docs]
- Protection: Transactions invisible to mempool searchers
- Adoption: Over $43 billion has transited through Flashbots Protect
- Criticism: Creates "dark pools" similar to traditional finance
Private RPCs
Many wallets and DEXs now offer private transaction submission that bypasses the public mempool. Examples: Metamask's "Private Transaction" feature, 1inch's Fusion mode.
Batch Auctions (CoWSwap)
Groups trades together and matches users directly when possible, eliminating ordering advantages. CoWSwap has processed ~$100 billion using this approach, with leading solver Barter handling $11B+.
MEV Recycling (Arbitrum TimeBoost)
Rather than fighting MEV, capture and redistribute it. Arbitrum's TimeBoost creates an "express lane" auctioned every 60 seconds—97% of revenue flows to the ARB DAO instead of concentrated with searchers.
MEV on Layer 2s
L2s face different MEV dynamics than Ethereum L1:
Base + Optimism: Cyclic arbitrage bots consume 50%+ of gas, pay less than 25% of fees
Arbitrum: 7% of gas from MEV bots (lower due to TimeBoost)
Problem: Between Nov 2024 - Feb 2025, Base added 11M gas/s capacity—almost all consumed by spam bots
Why L2 MEV Is Different
- No public mempool (usually): Sequencers order transactions privately
- Faster blocks: 250ms on Arbitrum, 2s on Base (vs 12s on Ethereum)
- Cheaper gas: Makes spam strategies viable
- Centralized sequencing: Single sequencer could extract MEV directly
L2 Solutions
- Flashblocks (Base, Unichain): 200ms mini-blocks provide near-instant confirmation
- TimeBoost (Arbitrum): MEV auction with revenue to DAO
- Rollup-Boost (Flashbots): Modular builder interface for OP Stack chains
How to Protect Yourself
Immediate Steps
- Use private RPC endpoints: Flashbots Protect, MEV Blocker, or wallet-provided options
- Set tight slippage: Lower slippage = less room for sandwich attacks (but higher revert risk)
- Use MEV-aware DEXs: CoWSwap, 1inch Fusion mode, Matcha
- Break up large trades: Smaller orders are less attractive targets
- Trade during low activity: Less competition for MEV during quiet periods
What Not to Do
- High slippage on large orders = guaranteed sandwich target
- Public mempool transactions for significant swaps
- Interacting with new token launches without protection (extremely high MEV)
The Economics of MEV
| Metric | Value |
|---|---|
| Total MEV extracted (2024) | ~$1.1 billion |
| Ethereum share | ~75% historically |
| Solana sandwich bots (30-day) | $4 million (50x Ethereum's $80k) |
| Flashbots Protect volume | $43 billion+ |
| L2 MEV (Polygon estimate) | $213 million lower bound |
The Bigger Picture
MEV is often described as crypto's "hidden tax"—value extracted from regular users by sophisticated actors with better technology and information access. Some key tensions:
- Efficiency vs fairness: Arbitrage improves markets; sandwiching harms users
- Decentralization vs protection: Private mempools protect users but centralize power
- Transparency vs dark pools: Public mempools enable MEV; private channels resemble TradFi
The ecosystem continues to evolve—expect ongoing innovation in both MEV extraction and MEV protection as billions of dollars remain at stake.