What is MEV and why does it matter?

When swapping tokens on a DEX, users usually focus on execution price, gas fees, and the number of tokens they ultimately receive. However, the outcome of a transaction also depends on a less visible factor known as block ordering.

A blockchain does not process all transactions at the same time. Instead, transactions are placed into a block in a specific order, and because each transaction leaves behind a new state, a later transaction must operate under the market conditions created by the one before it. When execution order materially affects the financial outcome for different parties, the ability to influence that order carries economic value. The value extracted from this process is known as MEV.

What is MEV?

MEV (Maximal Extractable Value) refers to the value that can be extracted during block production by selecting which transactions are included and determining their execution order within a block.

The term was previously known as Miner Extractable Value because Ethereum used Proof of Work (PoW), where miners were responsible for producing blocks. After The Merge, Ethereum transitioned to Proof of Stake (PoS), with validators replacing miners in the block production and transaction validation process. As a result, Maximal Extractable Value became the standard term.

This change in terminology also reflects a broader distinction. MEV is not tied to a specific participant, such as a miner. Instead, it arises from the way a blockchain orders and executes transactions, creating opportunities to influence transaction outcomes and capture additional value.

Why does transaction order have value?

The value of transaction placement stems from the constant rebalancing of decentralized liquidity pools, a dynamic driven by real-time price discrepancies across isolated protocols.

Suppose ETH is trading at $3,000 on DEX A and $3,020 on DEX B. An arbitrageur can buy ETH on DEX A and sell it on DEX B to capture the price difference. This opportunity exists only briefly. When the arbitrageur buys ETH on DEX A, the price there rises, while selling ETH on DEX B pushes the price lower. The two prices gradually converge, reducing the available arbitrage profit.

If several bots identify the same opportunity, the transaction executed first usually has the greatest advantage. Once the first bot completes the arbitrage, the state of both liquidity pools changes, meaning the bots that follow may capture only a small portion of the profit or may no longer have a profitable opportunity after fees.

How is MEV extracted?

The process of MEV extraction relies entirely on the structural mechanics of block production, execution queues, and the specific sequence that governs a transaction from inception:

• When a user submits a transaction, such as swapping ETH for USDC, it is usually not executed immediately. If the transaction goes through the public mempool, it remains in a waiting area alongside many other transactions before being included in a block.

• During this period, bots that specialize in finding MEV opportunities, commonly called searchers, monitor pending transactions and the blockchain's current state. Their goal is to identify situations where executing one transaction before or after another could generate a profit.

• For example, if a large swap is about to move the price on Uniswap, a searcher can calculate whether the resulting price change could create an arbitrage opportunity with another DEX. If a loan on Aave is approaching liquidation eligibility, a searcher can also prepare a transaction to liquidate the position before competing bots.

Identifying an opportunity is only the first step. The searcher's transaction must execute at the required position within the block for the strategy to remain profitable, making transaction placement a critical execution constraint. This requirement brings the block construction process into focus.

• An Ethereum block contains many transactions arranged in a specific order. Before a block is added to the blockchain, one party must select which transactions to include and determine their order. This process is commonly referred to as block construction.

• On Ethereum today, this work can be performed by specialized block builders. Builders receive transactions and groups of transactions submitted by searchers, then test different arrangements to construct a block with high value.

• When it is time to produce the next block, Ethereum selects a validator to act as the proposer. The proposer does not necessarily need to search for arbitrage or liquidation opportunities or manually arrange individual transactions. Instead, it can receive blocks prepared by builders and choose one block to propose to the network.

The process can be simplified into four steps.

• Users submit transactions

• Searchers identify MEV opportunities

• Builders sequence transactions into blocks

• Validators propose the selected block to Ethereum.

Not every MEV transaction follows this exact process because some order flow can move through private channels instead of the public mempool. However, this model helps explain why MEV today is no longer simply about one bot paying a higher gas fee to place its transaction ahead of another.

Common types of MEV

Following identification and block inclusion, MEV opportunities generally fall into three major categories. These mechanisms can either improve market efficiency or directly degrade execution quality for users.

Arbitrage

Arbitrage is one of the most common, widely used, and easiest forms of MEV to understand. In the DEX A and DEX B example above, ETH trades at different prices across the two markets. A searcher can buy ETH in the market where the price is lower and then sell it in the market where the price is higher to capture the difference.

This opportunity creates MEV because multiple searchers may identify the same price discrepancy, but they cannot all extract it. Once one searcher completes the arbitrage, prices across the two liquidity pools change, and the remaining price difference may no longer be large enough for other searchers to profit.

Liquidation

MEV also appears in lending protocols when a loan becomes eligible for liquidation. Protocols such as Aave generally require borrowers to maintain enough collateral relative to their debt. If the value of the collateral falls and the position crosses the required safety threshold, the protocol allows a liquidator to process part or all of the position and receive an incentive.

The economic value in this case comes from the liquidation incentive attached to restoring an unsafe lending position. Once a successful liquidation updates the loan state, the same opportunity is no longer available to other searchers.

Sandwich Attack

A sandwich attack extracts value differently from the two methods above by exploiting the expected price impact of a user's transaction. Suppose a user is preparing to buy a large amount of a token on an AMM. A searcher identifies that the swap could push the token price higher and places a buy order immediately before the user's transaction. The user's swap then moves the price higher, after which the searcher sells the tokens it previously purchased to capture the resulting price movement.

The transaction sequence consists of the searcher's buy order, the user's swap, and the searcher's sell order. Because the user's transaction is placed between the searcher's two orders, the strategy is called a sandwich attack.

A prominent example of MEV risk occurred in June 2026, when jaredfromsubway.eth, one of Ethereum's most active sandwich bots, was drained of approximately $7.5M. The scale of this loss reflects the bot's historically massive operational volume. Between February and May 2023, EigenPhi data shows the bot executed 238,000 sandwich attacks against 106,000 victims, generating $40.65M in gross revenue and $6.3M in net profit after spending $34.35M on gas fees.

The attacker did not steal private keys or exploit traditional smart contract bugs. Instead, they spent weeks deploying 66 fake token contracts and liquidity pools structured to mimic profitable MEV opportunities. When the bot interacted with these pools, it automatically granted token spend approvals for WETH, USDC, and USDT to attacker-controlled contracts. Because these approvals were not revoked post-transaction, the attacker drained the entire asset balance in a single transaction.

The event demonstrates how MEV bots can be weaponized against their own profit-seeking algorithms. If decision-making logic and token approvals are not strictly verified, machine-speed automation can rapidly amplify minor operational oversights into catastrophic capital losses.

The Impact of MEV

MEV is not inherently positive or negative because its impact depends on the source of the profit extracted by searchers and which party bears the cost during the process.

Positive Impact

Some forms of MEV help DeFi operate more efficiently. Arbitrage is a clear example because searchers profit from price differences across markets while simultaneously bringing prices on different DEXs closer together. This is particularly relevant for AMMs, where prices change through trading activity within liquidity pools and arbitrageurs help prices better reflect the broader market.

Liquidation also creates MEV while performing a necessary function for lending protocols. When a position becomes unsafe, incentives for liquidators help the system process the loan before it risks becoming bad debt.

Negative Impact

MEV becomes more concerning when a searcher's profit comes directly from degrading execution quality for users. At the user level, sandwich attacks can add costs to traders' orders. A searcher takes advantage of a large swap by trading before and after the user, which can leave the trader with a less favorable execution price. In this case, MEV becomes part of the execution cost even though it does not appear as a separate fee.

At the network level, MEV can also make the network less decentralized. Finding and extracting MEV efficiently requires specialized infrastructure, data, and software, which can give larger operators an advantage over smaller validators. If this advantage continues to accumulate, block construction could become concentrated among a small number of participants with the strongest ability to extract MEV.

Solutions to reduce the negative impact of MEV

Because MEV naturally emerges when transaction ordering affects economic outcomes, most current solutions do not attempt to eliminate MEV entirely. Instead, they focus on reducing harmful effects and improving how the extracted value is distributed. At the user level, one common approach is private transaction routing, where transactions are not immediately exposed through the public mempool. This makes it more difficult for searchers to observe an order before it is executed, reducing the risk of front-running and sandwich attacks. Flashbots Protect and several existing swap protection mechanisms follow this approach.

Another approach is intent and auction-based execution. Instead of submitting a transaction with a fixed execution path, users describe the outcome they want, while multiple parties compete to provide the best execution. Models such as UniswapX and CoW Protocol use competition between fillers or solvers to improve execution prices and reduce the likelihood that MEV value is extracted entirely from users.

At the network level on Ethereum, the focus is on how blocks are constructed. Mechanisms such as MEV-Boost and proposer-builder separation separate block optimization from the validator's responsibilities, meaning validators do not need to build complex MEV infrastructure themselves to compete for revenue. Ethereum is also continuing to research approaches such as inclusion lists and encrypted mempools to reduce builder power, improve censorship resistance, and limit the ability to exploit transaction information before execution.

Overall, current MEV solutions are developing along two parallel directions. Applications are working to protect execution and return more value to users, while Ethereum at the network level is focusing on block-building structure and the risk of network concentration. The goal is therefore not to make MEV disappear, but to limit harmful forms of MEV and control how this value is extracted across the ecosystem.

Conclusion

MEV is a natural consequence of transaction ordering in blockchains, creating both opportunities and risks for DeFi. Arbitrage and liquidation can improve market efficiency and support protocol operations, while harmful strategies such as sandwich attacks can worsen user execution.

The challenge is not to eliminate MEV, but to reduce its negative effects and ensure that value extraction does not come at the expense of users, fair access, or network decentralization. As Ethereum and DeFi continue to evolve, better transaction routing, execution mechanisms, and block-building designs will play an important role in achieving this balance.