Backrunning

What Is Backrunning?

Backrunning is a transaction ordering strategy in which a searcher (an automated bot or actor) monitors pending transactions and places their own transaction directly after a target transaction within the same block. The goal is to profit from the price displacement or state change that the target transaction creates.

For example, when a large trade on a decentralized exchange (DEX) pushes a token's price on that venue away from its price on other exchanges, a backrunner submits an arbitrage trade immediately afterward to capture the difference. The target transaction executes first at its expected price, and the backrunner profits only from the residual imbalance.

Backrunning is part of the broader MEV landscape, which encompasses all the value that block producers and searchers can extract by reordering, inserting, or censoring transactions within a block. Among MEV strategies, backrunning is widely considered the most benign because it provides a genuine market service without harming the target transaction's outcome.

How It Works

A backrunning operation follows a predictable sequence, from mempool monitoring through profit capture:

1. A searcher bot monitors the public mempool (the queue of unconfirmed transactions) for large or significant pending transactions, such as a substantial swap on a DEX
2. The bot calculates how the pending transaction will shift prices across liquidity pools and exchanges once it executes
3. If the resulting price imbalance creates a profitable arbitrage opportunity (for example, a token is now cheaper on Uniswap than on SushiSwap), the bot constructs a transaction to capture the difference
4. The bot submits its transaction as a bundle through infrastructure like Flashbots (on Ethereum) or Jito (on Solana), requesting placement directly after the target transaction in the same block
5. If included, the backrunner buys the underpriced asset on one venue and sells it on the overpriced venue, pocketing the spread minus gas fees and builder tips

A Concrete Example

Consider a trader selling 2,857 ETH for USDC on a single DEX. That sale pushes ETH's price down on that venue relative to other exchanges. A backrunning bot detects this pending transaction, constructs an arbitrage bundle, and places it immediately after the large sell. The bot buys discounted ETH on the impacted venue, sells it at the higher market price elsewhere, and profits from the spread. The original seller still receives the execution price they expected: only the residual imbalance is captured.

Bundle Submission

On Ethereum, backrunners rarely submit transactions to the public mempool. Instead, they use private transaction channels like Flashbots to submit bundles directly to block builders. A bundle is an ordered set of transactions that must be included atomically: either all execute in the specified order, or none are included. This guarantees the backrunner's transaction executes immediately after the target.

// Simplified Flashbots bundle structure
{
  "txs": [
    "0x...",  // target transaction (already in mempool)
    "0x..."   // backrunner's arbitrage transaction
  ],
  "blockNumber": "0x10A2B3C",
  "minTimestamp": 0,
  "maxTimestamp": 1720000000
}

On Solana, Jito bundles serve the same purpose. Over 95% of Solana's staked SOL flows through Jito-client validators, making it the dominant MEV infrastructure on that chain. Searchers pay tips to validators for priority bundle inclusion.

Backrunning vs. Front-Running vs. Sandwich Attacks

The three most common MEV strategies differ primarily in where the searcher's transaction is placed relative to the target:

Front-running is predatory: it places a transaction ahead of the target to move the price against them, then profits from their worsened execution. Sandwich attacks are double-sided: the attacker front-runs to inflate the price, lets the victim trade at the inflated price, then backruns to sell at the peak.

Backrunning, by contrast, is single-sided. The target's transaction executes first, unaffected. The backrunner only captures value from the state that already changed. This distinction is why researchers and protocol designers generally classify backrunning as "benign MEV."

Why Backrunning Matters for DeFi

Backrunning is not merely tolerated in decentralized finance: it performs a structurally necessary function. Without competitive backrunners, DeFi markets would be less efficient and lending protocols would be less safe.

Cross-DEX Price Alignment

When a large trade shifts the price on one AMM, backrunning bots immediately arbitrage the difference across venues, pushing the impacted pool's price back toward the true market rate. This constant arbitrage activity keeps AMM prices accurate and reflective of broader market conditions.

Liquidation Execution

Lending protocols like Aave and Compound rely on liquidation bots to close undercollateralized positions before they generate bad debt. These bots backrun price oracle updates: when an oracle reports a price drop that puts a position underwater, a liquidation bot executes immediately after to seize collateral and repay debt. Without competitive backrunning, protocols would accumulate toxic positions.

MEV Refunds

Infrastructure like Flashbots MEV-Share returns a portion of backrunning value to users whose transactions created the opportunity. By default, 90% of the backrun value is refunded to the original user's address. This model aligns incentives: users benefit from their own transactions' market impact rather than losing that value entirely to searchers.

Use Cases

• DEX arbitrage: the most common backrunning use case, capturing price discrepancies across decentralized exchanges after large trades shift prices on individual venues
• Liquidation: backrunning oracle price updates to execute liquidations on lending protocols before positions generate bad debt
• NFT minting: placing a transaction immediately after a collection reveal or whitelist event to mint high-value tokens
• Token listing sniping: executing a buy immediately after a new token is listed on a DEX and liquidity is added
• Protocol rebalancing: capturing arbitrage from protocol-initiated rebalancing operations, such as treasury diversification trades or stablecoin peg maintenance

Risks and Considerations

Gas Fee Escalation

When multiple searchers compete to backrun the same target transaction in the public mempool, they bid up gas fees in a priority gas auction. This competition increases transaction costs for all network users, not just the searchers. Private transaction submission through Flashbots and similar infrastructure mitigates this by moving the auction off-chain.

Centralization Pressure

Sophisticated MEV infrastructure creates advantages for well-capitalized searchers with low-latency connections to block builders. Over time, this can centralize MEV extraction among a small number of professional firms, reducing competition and potentially enabling collusion. On Ethereum, over 90% of blocks are produced through the MEV-Boost auction system, concentrating power among a handful of builders and relays.

Cross-Chain Complexity

As DeFi activity spans multiple chains and Layer 2 networks, cross-chain backrunning introduces latency and execution risk. Arbitrage between an L1 DEX and an L2 DEX requires bridging assets, which takes time and introduces bridge security risk. Atomic execution guarantees available within a single chain do not extend across chains.

Evolving Countermeasures

DEX designs continue evolving to minimize extractable value. Intent-based trading protocols like CoW Swap use batch auctions that eliminate ordering-dependent MEV entirely. Encrypted mempools using Trusted Execution Environments (TEEs) and research into Fully Homomorphic Encryption (FHE) aim to make transactions opaque until block inclusion, allowing "blind backrunning" that preserves arbitrage without revealing transaction details. For a deeper look at MEV dynamics and mitigation strategies, see the mempool economics research.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.