Short-Run MEV

The most common form of short-run MEV, where searchers exploit price discrepancies for the same asset across different decentralized exchanges (DEXs) within a single block. This includes:

• DEX-to-DEX arbitrage: Buying low on one AMM and selling high on another.
• CEX-DEX arbitrage: Capitalizing on price differences between centralized and decentralized venues.
• Cross-chain arbitrage: Using bridges to move assets between chains for profit, though this often involves multi-block risks.

Profiting from undercollateralized loans in lending protocols like Aave or Compound. Searchers run bots to monitor loan health and are the first to submit a liquidation transaction when a position's health factor falls below 1. The searcher repays part of the debt in exchange for a liquidation bonus (e.g., 5-10%) from the collateral, which is the source of MEV. Speed is critical, as it's a winner-takes-most event.

A predatory form of MEV that targets a visible user transaction in the mempool. The attacker front-runs the victim's trade by placing their own order first, and back-runs it with a closing order, effectively 'sandwiching' the victim. This manipulates the AMM's price to the attacker's advantage, costing the victim slippage and price impact. It is a direct transfer of value from the user to the searcher.

Short-run MEV is defined by its ephemeral nature. Opportunities exist for only a few seconds—the time between a transaction being broadcast to the mempool and its inclusion in a block. Success depends on:

• Latency: Physical proximity to block producers and high-speed infrastructure.
• Gas Auction Dynamics: Outbidding competitors via priority gas auctions (PGAs) to ensure transaction placement.
• Block Builder Selection: Reliance on builders who include the profitable bundle.

Almost all short-run MEV strategies are predicated on a public mempool, where pending transactions are visible before confirmation. This transparency allows searchers to:

• Simulate block state changes to identify profitable opportunities.
• Analyze transaction flow and pending orders.
• Construct complex bundles of transactions to capture value. The rise of private transaction channels (e.g., Flashbots Protect, Taichi Network) and encrypted mempools aims to reduce this surface area.

The effect of short-run MEV on the network is mixed:

• Positive: Arbitrage and liquidations provide economic efficiency by correcting prices and maintaining protocol solvency.
• Negative: Sandwich attacks are purely extractive, harming users. All forms contribute to network congestion and higher gas fees during auctions.
• Redistribution: MEV profits are distributed among searchers, block builders, and validators via tips, influencing validator revenue and centralization pressures.

The most common form of short-run MEV, where a searcher profits from price differences for the same asset across different decentralized exchanges (DEXs) within a single block. The searcher's bot buys the asset on the cheaper DEX and sells it on the more expensive one in an atomic transaction.

• Example: Buying ETH on Uniswap and selling it on SushiSwap in the same block when a large trade creates a temporary price discrepancy.
• Impact: Generally considered beneficial as it helps align prices across markets, but extracts value from regular traders as "latency tax."

A searcher repays a borrower's undercollateralized loan on a lending protocol (like Aave or Compound) to claim a liquidation bonus. Bots compete to be the first to submit this profitable transaction when a position's health factor falls below the threshold.

• Mechanism: The searcher supplies the owed assets, receives the borrower's collateral at a discount, and instantly sells it for profit.
• Role: While profitable for searchers, this activity is critical for protocol solvency by ensuring bad debt is cleared.

A predatory strategy where a searcher exploits a visible pending DEX swap in the mempool. The searcher places one transaction before the victim's trade (to buy the asset, driving the price up) and one after (to sell at the inflated price), "sandwiching" the victim.

• Result: The victim receives worse execution (slippage), and the searcher profits from the artificial price movement.
• Mitigation: Traders use private transaction relays or commit-reveal schemes to hide their intent.

An advanced form of reorg MEV where a miner or validator intentionally reorganizes the blockchain to steal profits from a previous block's arbitrage or liquidation. They revert the chain to a point before the profitable transaction, then include their own version of it.

• Prerequisite: Requires significant hashing power (PoW) or staking weight (PoS) to successfully orphan blocks.
• Impact: Undermines blockchain finality and creates a risk for protocols that assume instant settlement.

Profiting from inefficiencies and information asymmetry in NFT markets. Common strategies include:

• Floor Sweeping: Sniping multiple NFTs listed below market price in a single block after a sudden price drop.
• Trait Sniping: Using bots to instantly buy NFTs with rare, undervalued traits the moment they are listed.
• Bidding Games: Manipulating English auction mechanics by placing last-second bids or retracting bids to discourage competition.

A searcher exploits the latency between an on-chain oracle update (like Chainlink) and the execution of dependent contracts. By triggering a large trade on a DEX that the oracle uses for its price feed, the searcher can temporarily skew the reported price.

• Goal: To trigger or avoid liquidations on lending platforms, or to profit from derivatives contracts that settle based on the manipulated oracle price.
• Defense: Oracles use time-weighted average prices (TWAPs) and multiple data sources to resist such attacks.

Searchers are independent actors or bots that identify and execute profitable MEV opportunities. They compete by:

• Submitting bundles of transactions to block builders.
• Using sophisticated algorithms to spot arbitrage, liquidations, or other on-chain inefficiencies.
• Paying priority fees to validators to have their bundles included. Their profit is the difference between the value extracted and the cost of execution.

Block builders are specialized entities that construct the content of a block. They receive transaction bundles from searchers and public mempools, then optimize the block's ordering to maximize the total value (including their own fees). Builders submit the complete block proposal to validators, competing to have their block accepted. This role is central to the proposer-builder separation (PBS) architecture.

A transaction bundle is a set of transactions submitted by a searcher to a block builder with specific execution constraints. Key properties include:

• Atomicity: All transactions in the bundle execute, or none do.
• Ordering: The searcher specifies the exact sequence.
• Privacy: Often sent via private relay to prevent frontrunning. Bundles are the primary vehicle for capturing short-run MEV like arbitrage, where multiple dependent trades must succeed together.

Arbitrage is the most common source of short-run MEV, involving the simultaneous buying and selling of an asset across different markets to profit from price discrepancies. On-chain examples include:

• DEX Arbitrage: Exploiting price differences between Uniswap, SushiSwap, etc., often triggered by a large trade on one venue.
• Cross-Chain Arbitrage: Capitalizing on price gaps between assets on different blockchains (requires bridging).
• Funding Rate Arbitrage: Profiting from differences between perpetual futures contracts and their spot price.

Liquidations are a predictable form of short-run MEV in lending protocols (e.g., Aave, Compound). When a borrower's collateral value falls below a required threshold, their position becomes eligible for liquidation. Searchers compete to:

• Be the first to submit a liquidation transaction, repaying part of the debt in exchange for seized collateral at a discount.
• Pay a higher gas fee to win the race. This activity is considered 'welfare-improving' MEV as it maintains protocol solvency, though it extracts value from the liquidated user.