Crypto Gloss

Front-Running

Definition:

Front-running in DeFi refers to the practice of monitoring unconfirmed transactions in the public mempool and submitting a transaction designed to profit from knowledge of those pending transactions — typically by paying a higher gas fee to jump ahead in block ordering. The term comes from traditional finance, where brokers illegally executed trades for their own accounts ahead of known client orders. In DeFi, it is technically legal and enabled by the transparent, public nature of the Ethereum mempool. Front-running is a major subcategory of MEV (Miner/Maximal Extractable Value) and encompasses several distinct attack patterns.

How It Works

All unconfirmed transactions on Ethereum and other public blockchains sit in a publicly visible waiting area called the mempool. Before a transaction is included in a block, anyone can see its contents: what token is being bought, how much, through which pool.

A front-runner (typically a bot) watches the mempool for profitable opportunities and inserts its own transaction with a higher gas fee (Priority Fee / tip), causing block producers to include the front-runner’s transaction first.

The fundamental mechanism:

  1. Victim submits a swap transaction: Buy 100 ETH of Token X on Uniswap, max slippage 1%
  2. Front-runner bot detects the transaction in the mempool
  3. Bot calculates: if I buy Token X just before this transaction, the price will rise, and I can sell after the victim’s transaction completes for a profit
  4. Bot submits: Buy Token X (higher gas fee → executes first)
  5. Victim’s transaction executes at a worse price (the bot’s buy already moved the price)
  6. Bot immediately sells Token X (higher gas fee → executes right after victim)
  7. Bot profits from the price difference; victim suffers price impact beyond expected slippage

Types of Front-Running / MEV Extraction

Sandwich Attack (most common retail-visible attack)

The bot places one transaction immediately before and one immediately after the victim’s swap. The “buy before” pumps the price; the victim buys at the inflated price; the “sell after” completes the extraction. Victims experience worse execution than expected even within their slippage tolerance.

Generalized Front-Running

Instead of understanding the specific protocol mechanics, generalized front-runners simulate any arbitrary transaction, check if making an identical or similar transaction first would be profitable, and if so, copy and front-run it. This approach dominates in competitive mempool environments and is performed by sophisticated bots.

Displacement (Replacement Attack)

The front-runner submits an identical transaction with a higher fee, effectively “replacing” the victim’s transaction and stealing the opportunity (most relevant for bot-to-bot competition in arbitrage).

Back-Running

Instead of jumping ahead, the bot inserts a transaction immediately after a known transaction that will change market state — for example, backing a large trade to capture arbitrage across DEXs. Less harmful to users but part of the MEV taxonomy.

Time-Bandit Attacks (theoretical)

If a miner/validator could profit more from reorganizing recent blocks than from current block rewards, they might reorg the chain to capture MEV retroactively. Not practically observed on Ethereum post-merge but theoretically possible.

MEV and Front-Running Ecosystem

Front-running is formalized and institutionalized through MEV infrastructure:

  • MEV bots: Automated programs monitoring mempools and executing strategies within milliseconds
  • Flashbots / MEV-Boost: Infrastructure allowing searchers to submit bundles to block builders, bypassing the public mempool and reducing failed-transaction spam
  • Block builders: Entities that assemble blocks from MEV bundles, selecting the most profitable combination
  • Proposer-Builder Separation (PBS): Ethereum’s architecture separating block proposing from block building — partially addressing validator MEV capture but not eliminating front-running

History

  • Pre-2020 — Mempool sniping and arbitrage bots operate on early DeFi protocols (Bancor, EtherDelta), but at small scale.
  • 2020 — Phil Daian et al. publish “Flash Boys 2.0,” formally defining and analyzing MEV. The term enters mainstream crypto discourse.
  • 2020-2021 — Flashbots launched to reduce “MEV chaos” — the wasted gas and network congestion from bots competing in the public mempool. Successful but also formalizes and professionalize MEV extraction.
  • 2022 — Post-Merge Ethereum introduces MEV-Boost, which most validators now use. Front-running via private bundles becomes the dominant mode.
  • Ongoing — Sandwich attacks remain a persistent tax on retail DeFi traders. Researchers estimate hundreds of millions of dollars extracted annually via sandwiching alone.

Common Misconceptions

“Front-running is hacking/illegal in DeFi.”

Front-running using public mempool information is not illegal in DeFi — it is a consequence of the transparent, permission-less design. Unlike traditional finance, where brokers have fiduciary duties to clients, DeFi has no such relationship. However, it is widely considered predatory and has driven significant effort toward mitigation.

“Using a high slippage tolerance protects you from sandwiching.”

Higher slippage tolerance actually makes you more vulnerable. A 5% slippage tolerance means the bot can extract up to 5% from your trade while still having your transaction succeed. Low slippage tolerance (e.g. 0.1-0.5%) limits what can be extracted but may cause transactions to fail if the market is volatile.

Defenses and Mitigations

  • Private mempools (Flashbots Protect, MEV Blocker): Submit transactions directly to block builders without going through the public mempool. Prevents most bot-based sandwiching.
  • Low slippage tolerance: Limits the maximum a sandwich bot can extract; may cause reverts in volatile markets.
  • Aggregators with MEV protection: 1inch Fusion, Cow Protocol, ParaSwap Delta offer intent-based or batch-based execution that provides MEV resistance.
  • CoW Protocol (Coincidence of Wants): Matches trades peer-to-peer where possible, bypassing AMM price impact and sandwiching entirely.
  • TWAP execution: Breaking large swaps into many smaller trades over time to avoid large, predictable single transactions.

Social Media Sentiment

Front-running is one of the most discussed DeFi UX problems on crypto Twitter/X and r/defi. “Sandwich bots ate my trade” is a common retail complaint. Security researchers post real-time MEV dashboards (EigenPhi, libMEV) that track sandwich attacks, generating significant engagement. The debate between “MEV is a necessary efficiency mechanism” (extractors, researchers) and “MEV is a predatory tax on retail users” (consumer advocates, aggregators) is unresolved. MEV Blocker, Flashbots Protect, and CoW Swap are frequently recommended as practical defenses.

Last updated: 2026-04

Related Terms

See Also

Sources