Slippage Tolerance Explained: Settings That Limit Sandwich Losses

Slippage tolerance is the maximum price deviation you will accept on a swap before it reverts. It is a safety limit, and it is also the ceiling on what a sandwich bot can take: the looser you set it, the more room a bot has to push the price against you.

Slippage, price impact and the quote

Three terms get mixed up:

• Quoted price: what the interface shows when you request the swap.
• Price impact: the movement your own trade causes by changing the pool’s balance. Bigger trades in smaller pools have larger impact.
• Slippage: the gap between the price you expected and what you got, including changes that happened between quote and inclusion in a block.

The tolerance you choose is applied to the minimum amount you will receive. Interfaces convert it into a “minimum received” number embedded in the transaction. If the pool cannot deliver at least that, the swap fails.

Why the setting matters for sandwiching

A sandwich bot looks at your minimum received and your trade size. The difference between the expected output and your minimum output is the room. The bot’s front-run uses it up, up to the limit, so your swap just clears. A wide setting is an invitation. This is why what a sandwich attack is always returns to this one parameter.

Key takeaway: slippage tolerance is not only protection against volatility. It is also the budget you give an attacker.

Typical situations

The ranges below are general rules of thumb, not recommendations for any specific trade, and conditions vary.

Situation	Reasonable approach	Why
Stablecoin to stablecoin in a deep pool	Very low tolerance	Little real price movement, so any larger setting is unnecessary exposure
Major token pair in a deep pool	Low tolerance	Quotes are stable and impact is small
Smaller token with moderate liquidity	Moderate, with a smaller trade size	Some volatility is real
New or taxed tokens	Higher tolerance may be needed	Also high exposure: consider avoiding or using protected routing
Large order of any kind	Split or use protected routing	Price impact alone creates headroom

The cost of setting it too low

If the tolerance is below the actual price movement, your transaction reverts. You pay network fees and get nothing. In fast markets, repeated failures can cost more than a modest slippage would have. The aim is the tightest setting that works in current conditions, not the lowest number possible. Adjust gradually and watch the quoted price impact.

Tokens that need more slippage

Some tokens charge a fee on transfer, and others have thin pools. A swap with a transfer tax delivers fewer tokens than the quote suggests, so the interface may need a higher tolerance to succeed. That is a red flag in two ways. You are paying a tax, and you are giving bots room. If you must trade such tokens, keep sizes small and consider protected routing as described in how to avoid sandwich attacks.

How bots read your setting

CryptoMEV Bot’s sandwich bot page says its software looks for large swaps with high slippage tolerance. That is an unusually plain statement of what makes a trade attractive, and it should inform your settings. Combine a high setting with a large trade in a thin pool, and you have the textbook target.

Settings in common interfaces

Interfaces differ. Some, like Uniswap’s, offer an automatic option that adjusts to conditions, and manual entry. Defaults can change over time, so check the current docs and your own wallet. On Solana, aggregator interfaces expose similar controls, and defaults for volatile tokens are often wide. See the Solana guide for context.

Checking after the fact

Compare the amount you received with your minimum. If it is within a hair of the minimum on a trade where the quote suggested better, a sandwich is plausible. How to check if you were sandwiched explains how to confirm it on a block explorer.

A simple checklist

1. Look at pool depth and your share of it.
2. Start with a low tolerance for the pair type.
3. If it fails, raise by small steps.
4. Use private routing where available.
5. Re-read the minimum received before signing.

For operators the same logic runs in reverse: exposed trades are those with generous tolerance, and the supply of them is shrinking as habits improve, which is covered in are sandwich bots profitable. For related background on the vendor’s wider product line, its article on high-frequency crypto trading is a vendor-side view.

Worked example of the trade-off

Consider two attempts at the same swap. In the first, you set a very tight tolerance. The quoted price moves slightly in the seconds before inclusion, the swap reverts and you lose the network fee. You try again at the same setting and it succeeds. In the second, you set a loose tolerance and the swap succeeds first time, but a bot used most of your headroom and you received close to your minimum. Which cost more depends on trade size and fee levels: for small trades the revert fee can dominate, for large trades the extraction can.

That is why a single universal setting does not exist, and why the advice is to size the tolerance to the trade. For related reading on how the mechanism works from the bot’s side, see the vendor’s sandwich bot page.

Common mistakes

1. Copying someone else’s setting. What worked for another pair, size or day may not work for yours.
2. Raising slippage after one failed attempt without checking why. A revert may reflect a transfer tax, a thin pool or a fast move, and each needs a different response.
3. Ignoring price impact. The interface shows it for a reason. If impact alone is large, slippage tuning does not fix the problem; trade size and venue do.
4. Leaving a wide default on after a volatile token trade. Reset it for your next swap.

These are small habits, but they compound across many swaps.

This article is educational and not financial advice.