You open Uniswap. You want to trade 100 ETH. The price shows $2,500 per coin. You set 2% slippage tolerance (standard practice) and hit swap.

You think: "I'll get roughly $250,000 in return."

What actually happens: A bot detects your transaction in the mempool before it gets confirmed. The bot immediately buys the same token you're about to buy, driving the price to $2,550. Your transaction goes through at the inflated price. Then the bot sells its position back into your trade, pocketing the profit. The price crashes back to $2,500.

You end up spending $255,000 instead of $250,000. You lost $5,000 on a transaction you thought was safe.

The bot made $3,500. Block builders made the rest.

This is called a "sandwich attack," and it's one of the reasons crypto hasn't gone mainstream: ordinary people keep losing money to invisible enemies they never knew existed.

The Scale? $1.1 Billion in the Gray Market

Let's get the scope right, because the numbers are apocalyptic:

According to ZENMEV data, MEV on global blockchains has skyrocketed from under $100 million in 2020 to over $1.1 billion by the end of 2024. That's an eleven-fold increase in five years.

On Ethereum specifically, sandwich attacks alone account for roughly 66% of all MEV activity. On Solana, sandwich bots extracted between $370 million to $500 million over just a 16-month period (January 2024 to May 2025).

But here's what really matters: Cointelegraph's exclusive data from EigenPhi, covering more than 95,000 sandwich attacks from November 2024 to October 2025, shows that while attacks result in about $60 million in annual losses for traders, the actual value extracted is far larger. Block builders capture most of the value through gas fees, meaning your loss is spread across multiple profiteers.

How the Attack Actually Works

A sandwich attack has three parts:

Front-run: The bot detects your pending transaction in the public mempool (every blockchain broadcasts pending transactions before they're confirmed). The bot quickly buys the same token you're about to buy, driving the price up.
Victim transaction: Your trade executes at the artificially high price. You get less output than you expected.
Back-run: The bot sells its position, locking in profit. You've been sandwiched. The bot made money. You lost money.

It takes milliseconds. You never see it coming.

The Flashbots research team calls this the "invisible tax"—profit that miners and validators collect from other participants without most users even knowing it exists.

The Most Infamous Example: jaredfromsubway.eth

In 2023-2024, a bot known as jaredfromsubway.eth became infamous for a singular reason: It spent millions of dollars in gas fees just to execute sandwich attacks.

Here's the insanity: The bot would intentionally bid up gas prices—sometimes by 10-20x normal levels—just to cut in line and position its transactions before and after yours. It was profitable even while hemorrhaging money on gas fees.

This tells you how lucrative sandwich attacks are. A bot will spend millions to make more millions.

The bot's strategy: target high-slippage trades (especially memecoins where people set 10-20% slippage tolerance). Use brute force gas bidding to cut in line. Extract the spread. Repeat thousands of times per day.

According to Outlook India, jaredfromsubway.eth extracted significant value throughout 2023-2024, despite the astronomical gas costs, proving that sandwich attacks remain profitable even under the most hostile conditions.

The Reality in 2025? It's Getting Worse, Not Better!

Here's the hopeful news: Sandwich attack profitability has actually declined in 2025.

According to Cointelegraph's October 2025 data, monthly sandwich extraction fell from nearly $10 million in late 2024 to about $2.5 million by October 2025. Average profit per sandwich attack dropped to just $3.

So the bots are disappearing, right?

Wrong.

The number of bots has stayed relatively constant (515 distinct bots on Ethereum in October 2025), but they've gotten more aggressive. Instead of targeting huge trades, they now execute thousands of small-scale sandwich attacks per day.

Jared, the most profitable sandwich bot in 2025, targeted a token pair thousands of times across 2025, extracting nearly $19,000 across 65 sandwich attacks—averaging roughly $292 per attack.

The economics haven't changed. The strategy has: quantity over quality.

The Casual Cruelty of the System

Here's what makes this infuriating: About 38% of sandwich attacks target low-volatility pools (stablecoins, wrapped assets, liquid staking tokens). In other words, people who think they're trading "safe" assets end up getting hit hardest.

Roughly 12% of all sandwich attacks hit stable swaps—trading between stablecoins like USDC and USDT. You're trying to swap one dollar for another dollar, and a bot extracts slippage from you anyway.

This is especially damaging because traders don't expect it. They set low slippage tolerance on stablecoin swaps because they think the price will barely move. Then a sandwich attack forces their slippage to 5%, 10%, or higher without them realizing.

Why This Is So Hard to Detect

Most users never realize it's happening.

You swap 100 USDT for USDC. You get 98.5 USDC instead of 99.8. You think "prices just move, I guess" and move on with your life.

You have no idea a bot just extracted $1.30 from your $100 trade.

Multiply that across millions of users, thousands of trades per day, and you have a silent wealth extraction machine.

The Solana Specifics: Even Worse

On Solana, the situation is actually worse. Sandwich bots pay only 15-20% of their profits to validators as tips (compared to arbitrage bots that pay 50-60%). This means sandwich bots get to keep most of their stolen value.

Over 16 months (January 2024 to May 2025), Solana sandwich bots extracted $370-500 million. That's roughly $23-31 million per month—on a single blockchain.

And Solana's mempool was shut down in March 2024 to prevent sandwich attacks. It still happened. The attacks just migrated to alternative mempools.

What Can You Actually Do?

Here's the uncomfortable truth: If you're an ordinary trader, there's almost nothing you can do to completely prevent this.

You can:

Use private order flow routers (send transactions directly to validators instead of the public mempool)
Keep slippage low (though this isn't foolproof; bots still attack low-volatility pools)
Use MEV-protected wallets (like CoolWallet, which offers MEV protection)
Avoid high-liquidity pools during gas wars (when gas prices spike 10-20x, it's a sign sandwich attacks are happening)

But honestly? These are band-aids on a systemic problem.

The real issue is architectural: As long as transactions are transparent before confirmation, bots can see them coming.

The Broader Problem: Protocol-Level Failure

Ethereum has attempted to address this with Proposer-Builder Separation (PBS), implemented in 2024. The idea is to separate the roles of transaction orderer (builder) and block proposer (validator) to reduce MEV concentration.

Solana shut down its public mempool in March 2024—a nuclear option that reduced but didn't eliminate sandwich attacks.

Other chains like Shutter and rollups are exploring threshold encryption (making transactions invisible until after they're confirmed).

But here's the kicker: Nobody has solved this. We're 15 years into cryptocurrency and MEV extraction remains one of the most persistent problems in the ecosystem.

Why This Matters Beyond Crypto

This isn't just a crypto problem. This is a market infrastructure problem.

Traditional finance has been fighting order flow exploitation for 40 years. Stock exchanges have massive regulatory frameworks around front-running. It's illegal. People go to jail.

In crypto, it's legal. It's baked into the protocol. And it generates $1.1 billion per year.

We've built a financial system where the infrastructure itself enables and profits from exploiting ordinary users. And most people don't even know it's happening.

The Bottom Line

Every time you trade on a decentralized exchange without MEV protection, you're gambling that a bot doesn't notice.

Sometimes you win. Most of the time, if you're not paying attention to slippage, you're subsidizing the MEV economy.

Until protocols implement native MEV protection at the base layer, sandwich attacks will remain a tax on everyone who trades crypto.

The only question is: How long are we willing to pay it?

References & Sources

MEV Scale & Historical Growth:

ZENMEV Data: MEV growth from <$100M (2020) to $1.1B+ (end of 2024): https://www.coolwallet.io/blogs/blog/mev-protection
Flashbots Research: MEV as "invisible tax" on blockchain participants: https://info.arkm.com/research/beginners-guide-to-mev

Ethereum Sandwich Attacks:

Cointelegraph Exclusive Data from EigenPhi (Dec 4, 2025): 95,000+ sandwich attacks analyzed (Nov 2024 - Oct 2025): https://www.tradingview.com/news/cointelegraph:fa12ba092094b:0-exclusive-data-from-eigenphi-reveals-that-sandwich-attacks-on-ethereum-have-waned/
Ethereum sandwich attacks = 66% of total MEV activity: CoolWallet (June 2025)
Annual losses to traders: ~$60 million (EigenPhi data, Oct 2025)
Monthly extraction trends: ~$10M (late 2024) → $2.5M (Oct 2025)
Average profit per attack: $3 (Oct 2025)

Solana MEV Data:

Solana Compass (Accelerate 2025): Sandwich bots extracted $370-500M over 16 months (Jan 2024 - May 2025): https://solanacompass.com/learn/accelerate-25/scale-or-die-at-accelerate-2025-the-state-of-solana-mev
Helius Solana MEV Report (Aug 1, 2025): Memecoin traders especially vulnerable; high slippage tolerance targets: https://www.helius.dev/blog/solana-mev-report
Sandwich bot compensation: 15-20% to validators (vs. 50-60% for arbitrage bots)
Jito public mempool shutdown: March 2024 (Solana Compass)

Jaredfromsubway.eth Case Study:

Outlook India: Bot spent millions in gas fees to execute sandwich attacks across Uniswap (2023-2024): https://www.outlookindia.com/xhub/blockchain-insights/what-are-sandwich-attacks-how-mev-bots-drain-millions-from-crypto-traders
Jared Strategy 2025: Targeted 65 attacks on MANYU/WETH pair, extracted ~$19,000 (Cointelegraph, Oct 2025)
High gas bidding mechanics: 10-20x normal levels during MEV competition (Arkm, ARKM Research)

Technical Mechanics & Pool Targeting:

Sandwich attack three-part structure: Front-run → Victim → Back-run (CoolWallet, Arkm)
38% of attacks target low-volatility pools (stablecoins, LSTs): EigenPhi data (Oct 2025)
12% hit stable swaps (USDC/USDT): Creates unexpected slippage (EigenPhi, Oct 2025)
Gas wars: Spikes 10-20x during high-profit MEV opportunities (Arkm Research, Sept 2024)

2025 Bot Competition Data:

Ethereum sandwich bots in Oct 2025: 515 distinct bots, ~100 active per month (EigenPhi, Dec 2025)
Profitability distribution: 1/3 operating near breakeven; 30% recording net losses (EigenPhi, Oct 2025)
Only 6 bots generated >$10,000 total profit in 2025 (EigenPhi, Dec 2025)

Protocol-Level Solutions:

Ethereum's Proposer-Builder Separation (PBS): Implemented 2024 to reduce MEV concentration: Arkm Research
Solana Public Mempool Shutdown: March 2024, reducing but not eliminating attacks: Solana Compass
Shutter Network & Threshold Encryption: Proposed solutions for transaction privacy: Cointelegraph Research (Dec 2025)
CoWSwap Batched Processing: Alternative transaction ordering to prevent MEV: Arkm Research

User Protection Tools:

MEV-Protected Wallets: CoolWallet and others offering built-in MEV protection (2025): https://www.coolwallet.io/blogs/blog/mev-protection
Private Order Flow Routers: Sending transactions to validators instead of public mempool
sandwich.me: Real-time tracking and analysis tool for Solana MEV (launched June 2024, enhanced versions available): https://www.solanacompass.com/

Home

Untitled Page 1

Slippage in Crypto: Why You're Paying More Than You Think (And MEV Bots Are Stealing From You)