# backrunning extracts value from state changes without harming the preceding transaction Backrunning inserts a transaction immediately after a target transaction to capture value from the state change the target creates. Unlike frontrunning and sandwiching, backrunning does not worsen the target user's execution — the user's trade completes first, and the backrunner profits from the resulting price discrepancy. This makes backrunning ethically and economically distinct from sandwich attacks. The two primary backrunning use cases are arbitrage and liquidations. Arbitrage backrunning restores price equilibrium across venues after a large trade moves one pool's price — a net benefit to market efficiency. Liquidation backrunning keeps lending protocols solvent by competing to clear underwater positions before they accumulate as bad debt. Since [[bad debt accumulation in lending protocols occurs when liquidations fail to clear underwater positions during extreme market events]], liquidation MEV competition directly serves protocol health by ensuring liquidators remain incentivized. Competition for backrunning in the public mempool creates priority gas auctions (PGAs): multiple bots bidding escalating gas fees to ensure their backrun lands first after the target transaction. These gas wars generate network congestion and burn ETH — externalities that affect all users even though no individual user is directly harmed by the backrun itself. Speculative MEV extends the backrunning concept to L2s. On chains with very low transaction costs, searchers submit many backrun attempts that revert on failure, because the cost of a reverted transaction is minimal relative to the potential gain. Since [[revert-based MEV on L2s exploits cheap failed transactions to make speculative backrunning strategies profitable]] (a pattern documented on rollups), this creates a new MEV class specific to L2 economics: high-volume speculation rather than the targeted, profitable execution that characterizes mainnet backrunning. MEV-Share was designed specifically to return backrun profits to users: searchers submit partial bundles that backrun MEV-Share transactions, and 90% of generated MEV flows back to the originating user. This makes backrunning a redistribution mechanism rather than pure extraction. --- Relevant Notes: - [[sandwich attacks exploit AMM deterministic pricing to front-run and back-run victim trades within a single block]] — the harmful contrast: sandwiching combines frontrunning and backrunning to worsen user execution, unlike standalone backrunning - [[bad debt accumulation in lending protocols occurs when liquidations fail to clear underwater positions during extreme market events]] — liquidation backrunning competition prevents this failure mode - [[frontrunning exploits public mempool visibility to insert competing transactions before profitable pending operations]] — the harmful frontrunning that backrunning is often contrasted with in MEV discussions Topics: - [[vulnerability-patterns]] - [[protocol-mechanics]]