MEV, short for maximal extractable value, is the additional profit that can be captured by including, excluding or reordering transactions within a block, beyond the standard block reward and gas fees. In smart contracts on public blockchains, MEV arises because transaction order is decided by whoever builds the block, and many contract outcomes, such as a swap price or a liquidation, depend directly on that order.
The term originally stood for miner extractable value under proof of work. After Ethereum moved to proof of stake, the name changed to maximal extractable value, because the opportunity now belongs to validators, block builders and searchers rather than miners. For anyone working in crypto & blockchain, MEV is a structural property of open transaction pools, not a bug in a single contract.
How MEV is created and captured
MEV is created whenever the state change of one transaction alters the value of another. A pending trade on an automated market maker moves the pool price, a falling oracle price makes a loan eligible for liquidation, and two exchanges briefly quote different prices for the same token. Each event leaves a profit opportunity for whoever controls ordering in the next block.
Pending transactions usually wait in the public mempool, where anyone can read them before confirmation. Specialized bots called searchers simulate these transactions, find profitable sequences and submit bundles of their own transactions positioned before or after the target. Searchers then compete by paying a large share of the profit to the block producer as a priority fee or direct payment.
The value chain has several distinct roles. The table below shows who participates in MEV extraction on Ethereum and what each role contributes.
| Role | Main activity | Source of income |
|---|---|---|
| Searcher | Detects opportunities and builds bundles | Arbitrage or liquidation profit |
| Block builder | Assembles the most valuable block | Margin between bids and payment to proposer |
| Relay | Passes blocks from builders to proposers | Usually none or service agreements |
| Validator (proposer) | Signs and proposes the chosen block | Builder payment plus consensus rewards |
| User | Submits the original transaction | Bears the cost through worse execution |
Main types of MEV strategies
MEV strategies differ sharply in their effect on users. Some activity is broadly neutral or beneficial, because it keeps prices aligned and loans solvent, while other strategies transfer value directly from ordinary traders to the extractor.
The following table compares the most common strategies, how they work and how they affect other participants.
| Strategy | Mechanism | Effect on users |
|---|---|---|
| DEX arbitrage | Buys on one exchange and sells on another in one block | Aligns prices, mostly neutral |
| Liquidation | Repays an undercollateralized loan for a bonus | Keeps lending protocols solvent |
| Sandwich attack | Trades before and after a victim swap | Victim receives a worse price |
| Front-running | Copies or preempts a profitable pending transaction | Original sender loses the opportunity |
| Back-running | Trades immediately after a large price move | Usually neutral |
| Time-bandit attack | Reorganizes past blocks to capture old MEV | Threatens consensus stability |
The sandwich attack is the most visible harmful form. A searcher sees a large swap with a generous slippage tolerance, buys the same asset first to push the price up, lets the victim trade at that inflated price and then sells. The profit equals roughly the price impact the victim absorbed, minus gas and builder payments.
MEV compared with related concepts
MEV is a broader category than several terms often used as synonyms. Front-running is one specific technique within MEV, and gas fees are the competitive price paid to secure ordering rather than the value itself. The comparison below clarifies these boundaries.
| Concept | Scope | Relation to MEV |
|---|---|---|
| MEV | All value from transaction ordering | The overall category |
| Front-running | Acting before a known pending transaction | One MEV technique |
| Gas fee | Payment for computation and inclusion | Tool used to bid for position |
| Slippage | Difference between expected and executed price | The margin sandwich attacks exploit |
| Reentrancy attack | Exploit of flawed contract logic | Not MEV, because it needs a code bug |
A useful rule is that MEV requires no flaw in contract code. It exploits public information and ordering rights, so even a perfectly audited contract can leak value if its design exposes predictable price movements to observers.
Infrastructure that shapes MEV today
Before 2020, searchers competed through priority gas auctions, repeatedly raising fees in the public mempool. These auctions congested the network and wasted block space on failed transactions. The response was off-chain auction infrastructure that lets searchers bid privately for inclusion.
On Ethereum, proposer-builder separation is implemented today mostly through the out-of-protocol software MEV-Boost. Validators outsource block construction to a competitive builder market and simply choose the highest-paying block. This design spreads MEV revenue to all validators, including small ones, but it has concentrated block building among a small number of builders, which raises censorship and centralization concerns.
Other mitigations target users directly. Private transaction relays hide trades from the public mempool, MEV-share schemes return part of the back-running profit to the user, and batch auctions settle many orders at a single uniform price so that ordering inside the batch has no value. Research into encrypted mempools and enshrined proposer-builder separation continues, and none of these approaches eliminates MEV completely.
Designing contracts and transactions to limit MEV
Developers and users can reduce exposure, although some MEV is unavoidable wherever ordering matters. The decision table below summarizes practical responses to typical situations.
| Situation | Recommended measure | Reason |
|---|---|---|
| Large swap on an AMM | Tight slippage limit and private relay | Removes the sandwich margin |
| Order-sensitive protocol logic | Commit-reveal or batch settlement | Hides intent until ordering is fixed |
| Oracle-dependent liquidations | Auction liquidation bonuses | Returns value to the protocol |
| Token launch or mint | Fair ordering or randomized allocation | Prevents bot capture of supply |
| Price-sensitive governance action | Time delay and announcement | Reduces surprise arbitrage |
A common mistake is setting a very high slippage tolerance to avoid failed transactions, which effectively invites a sandwich attack. Another is assuming that a layer 2 network is immune: rollups with a centralized sequencer shift ordering power to that sequencer, so the MEV risk changes form rather than disappearing.
In everyday language
Imagine a public notice board where everyone pins their orders before a clerk processes them in a batch. The clerk, or anyone who can pay the clerk, can read the orders and decide which one goes first. MEV is the money made from that power to choose the order.
Sometimes that power does useful work, such as fixing a price difference between two shops. Sometimes it hurts people, for example when someone sees your large purchase, buys just before you and sells to you at a higher price. The key lesson is that on a public blockchain, a pending transaction is visible, and visible intentions can be exploited.
Seen in the real world
A trader on Ethereum submits a swap of 200,000 stablecoins for ether on a popular automated market maker, with the wallet’s slippage tolerance set to 3 percent. A searcher bot detects the transaction in the public mempool within milliseconds and builds a sandwich bundle: a purchase of ether before the trade and a sale right after it.
The builder includes the bundle because the searcher offers most of its profit as payment. The trader still receives ether, but at a price about 1.5 percent worse than quoted, a loss of roughly 3,000 dollars. On the next large swap, the trader routes the order through a private relay with a 0.3 percent limit, and the trade settles at the quoted price with no sandwich exposure.