Sandwich Attack (MEV)

Sandwich attacks are one of the most prevalent forms of harmful MEV in DeFi, directly extracting value from regular users' trades. Understanding how they work is essential for any DeFi participant.

**The mechanics step by step:**

Suppose a user submits a trade to buy ETH with USDC on Uniswap, set to accept up to 1% slippage:

1. **Detection:** A sandwich bot monitors the mempool and spots the pending transaction 2. **Front-run:** The bot submits a buy order for ETH with a higher gas tip, ensuring it executes before the user's trade — this pushes the ETH price up 3. **Victim's trade executes:** Now the user buys ETH at the inflated price (within their 1% slippage tolerance) 4. **Back-run:** The bot immediately sells the ETH it just bought into the now-reduced liquidity after the user's purchase — at a profit

The bot profits from the price impact it caused and passed onto the victim.

**How much does this cost users?** Estimates suggest sandwich attacks extract hundreds of millions of dollars per year from DeFi users. Each individual attack may cost a user 0.1–2% of their trade value, which feels small but aggregates to significant losses for the ecosystem.

**Defenses:**

**1. Low slippage tolerance:** Setting tight slippage (0.1–0.3%) makes sandwich attacks less profitable — the bot's front-run may cause too much price impact, pushing the victim's trade beyond tolerance and reverting it. However, this causes more failed transactions in volatile conditions.

**2. Private mempool / MEV protection:** Services like Flashbots Protect, MEV Blocker, or 1inch Fusion route transactions through private channels where sandwich bots can't see them before inclusion. This is the most effective defense.

**3. DEX aggregators with MEV protection:** 1inch Fusion and CoW Protocol use off-chain matching that makes traditional sandwich attacks impossible.

**4. Limit orders:** Filling via limit orders rather than market swaps eliminates sandwich risk entirely.