Clicked buy at $100. Paid $103.

That $3 is slippage. Gone before you even knew the trade filled.

Happens on every market. Crypto worse than most. Thin tokens worse than liquid ones. Panic moments worse than calm ones. Just how it works.

Two Ways It Gets You

Price moves while your transaction is pending. Ethereum block takes 12-15 seconds to confirm. ETH moves 0.8% in that window. You pay 0.8% more than the price you saw. Didn't do anything wrong. Market just moved.

Or your order is big enough to move the price itself. Buying $200,000 of a token sitting in a $500,000 pool. First $20,000 fills close to your expected price. Last $20,000 fills significantly higher because your own buying pushed it there. Average fill price across the whole order is ugly.

Second one is entirely your fault for sizing too large relative to available liquidity. Not a criticism. Just how AMMs work.

AMMs Make This Messier

Order books show exactly where liquidity sits. Slippage calculable before you submit. At least you can see what you're walking into.

AMMs don't work that way. Uniswap, Raydium, whatever. Constant product formula. x times y equals k. Pool has to stay balanced. You buy, ratio shifts, price moves. Bigger the buy relative to pool size, harder the price moves against you before it finishes.

$5,000 into a $2 million pool. Barely registers. Maybe 0.1% slippage.

$5,000 into a $30,000 pool. Brutal. Could be 15% or worse. You moved the market with your own trade before it even finished.

Pool size is everything. Tiny pool, any meaningful order causes serious slippage. This is why low liquidity tokens are dangerous to trade in size.

The Slippage Tolerance Problem

Every DEX asks you to set one. Maximum acceptable gap between expected and actual price. Transaction reverts if slippage exceeds it.

Set it too tight. Token moves 3% while pending. Tolerance at 0.5%. Reverts. Gas fee burned. Nothing in your wallet. Try again.

Set it too loose. Bots notice. Sandwich time.

Rough guide that most traders use:

Stablecoins: 0.1%. Price shouldn't move at all. ETH, SOL, large caps: 0.5-1%. Volatile altcoins: 3-5%. Fresh Pump.fun launch with thin liquidity: sometimes 10%+. Also maximum sandwich risk. Pick your poison.

Sandwich Attacks

This is where high slippage tolerance gets genuinely expensive.

Transaction hits the mempool. Public. Visible. Bot sees it. Sees your 10% tolerance. Calculates profit.

Bot buys ahead of you. Pushes price up. Your transaction fills at the higher price. Within your stated tolerance so it goes through. Bot immediately sells into your buy. Pockets the difference. Done in milliseconds.

You paid the worst price you said you'd accept. Bot took most of that margin as pure profit. Happened before you could blink.

Defense: private RPC endpoints that route around the public mempool. Flashbots Protect on Ethereum. Jito on Solana. Some wallets build this in now. Not perfect but cuts sandwich exposure significantly. Worth using on any meaningful size trade.

Positive Slippage

Real thing. Nobody talks about it.

Market drops slightly between submitting and confirming. Expected to pay $100. Got it at $97.50.

Happens during fast selloffs. Submit a buy, price keeps falling for a second. Confirm. Better fill than expected.

Some protocols historically kept positive slippage as revenue instead of passing it back. Most return it to the user now. Worth knowing it exists even if negative slippage is the painful one everyone remembers.

Slippage FAQ

Is slippage the same as fees?

No. Fees are fixed. Slippage is variable price impact. Both eat into your return. Different mechanisms. Real cost of a trade is fees plus slippage combined. Most people only think about fees.