# What Is Slippage in Crypto? MEV, Price Impact & Safer Swaps

![Photorealistic trading workspace with an ETH/USDC decentralized-exchange quote, side-by-side price-impact and slippage-risk metrics, a liquidity-depth chart, and a discreet Terenval Wallet-branded smartphone; illustrative analytics, not a claim of a built-in swap or MEV-protection function.](hero.webp)

**Crypto slippage is the difference between an expected transaction result and the actual result received when a trade executes.** On decentralized exchanges (DEXs), this difference may arise while a quote waits for inclusion on-chain, because the market or pool reserves have moved. It is distinct from **price impact**—the price movement or execution penalty already caused by the size of the trade relative to available liquidity. Swap fees, network gas, price spread and potentially harmful maximal extractable value (MEV) are additional costs or mechanisms, not synonyms for the same percentage.

Imagine entering an ETH/USDC swap and seeing **3.9486 ETH** quoted for **10,000 USDC**. The quote may already account for the pool fee and the price impact of buying against its reserves. If you then set a **0.5% slippage tolerance**, the router may require at least about **3.9289 ETH**, rather than promising the original 3.9486 ETH. If conditions worsen beyond the permitted minimum before the transaction executes, a conventional on-chain exact-input swap should revert instead of delivering less than the minimum. A revert can still cost network gas. A high tolerance does not cause the price impact already in the quote; it permits a wider range of *additional execution deterioration*.

> **Short answer:** Slippage is the gap between your **quote** and **execution**; price impact is the cost of moving through available liquidity with your own trade. For a typical direct exact-input DEX swap, the interface estimates how many output tokens you should receive, then uses your slippage tolerance to calculate a **minimum acceptable output**. Tightening tolerance limits the downside of a successful execution but increases the chance of a failed trade. MEV sandwich attacks may worsen execution inside that permitted band. The best protection is to compare an independent reference price, inspect route liquidity and fees, and sign only a transaction whose minimum output, spender and network you understand.

**Terenval's editorial view:** *The right question is not “Which slippage percentage is best?” It is “How many final tokens can I receive, on what route, at what total cost, and with what enforceable downside limit?”* A wallet user who only reduces the tolerance slider may be overlooking an already-unfavorable quote, an opaque router fee, a shallow pool, a malicious token or a sandwichable public transaction. This guide presents five original editorial tools—the **Terenval Quote-to-Receipt Ledger**, **Depth Stress Grid**, **Slippage Exposure Budget**, **MEV Exposure Gate** and **Route Decision Framework**—to make those hidden trade-offs visible. They are educational models, not software features, deployed trading defenses or financial advice.

**Information cutoff: November 7, 2026.** Protocol mechanics are based on the official documentation cited throughout. All trading pairs, reserves, token prices, fees, network costs, quotes and MEV scenarios are **hypothetical teaching examples**, not live exchange offers or a historical measurement of Terenval users. Actual Uniswap interfaces, fee settings, supported routes, order mechanisms and protection services can vary by chain, token, version and date; inspect the current transaction details before signing.

## Key Takeaways

- **Slippage and price impact are different.** Impact comes from your trade's interaction with liquidity; slippage compares the quoted result with the actual fill, often after other market activity.
- **The quote can already include expected price impact and swap fees.** Do not mechanically add those same costs a second time when evaluating the result.
- **Slippage tolerance is a permission, not a prediction.** A 1% tolerance does not mean you will definitely lose 1% or that the market normally moves 1%.
- **A tighter limit may protect an execution but increase failed transactions.** Direct on-chain reverts can still consume gas.
- **A wider limit may permit a much worse fill.** A trade can succeed at a poor price even if your selected percentage was accepted by the interface.
- **Minimum received matters more than an isolated percentage.** For an exact-input trade, check the actual minimum number of tokens; for an exact-output trade, check the maximum tokens you might spend.
- **Trade size relative to liquidity depth is critical.** A $10,000 order can be small in a deep pool and disastrous in a shallow one.
- **Fees have multiple layers.** Pool/LP fees, protocol/hook fees, aggregator or solver fees, network fees and bridge costs must not be confused with slippage.
- **An external reference price is essential.** A pool can show low quote-to-fill slippage while its quoted rate is already far from the wider market.
- **MEV is not synonymous with all arbitrage.** Reordering and sandwiching can harm swappers, while some arbitrage helps align markets; transaction attribution requires evidence.
- **Public order visibility may expose trades to sandwiches.** Private routing and auction-based execution can mitigate some risks but introduce trust, availability and settlement assumptions.
- **UniswapX is not a normal slippage-slider swap.** Its order terms and minimum output matter; the ordinary Uniswap interface slippage setting does not apply to UniswapX orders.
- **Uniswap v4 can have custom fee and hook behavior.** Do not assume every pool has the same invariant, quote protections or contract risk.
- **Layer 2 does not abolish price impact or MEV.** Lower network fees do not guarantee deeper liquidity or better execution prices.
- **Self-custody is authorization, not execution insurance.** A signed transaction can be valid yet economically unfavorable; verify spender, tokens, network and minimum output.

## What Is Slippage in Crypto? A Plain-English Explanation

**Slippage is the difference between the expected execution terms of a trade and the terms actually executed.** The term applies to centralized order books and decentralized liquidity pools, although the causes and mechanics differ. On a DEX, the relevant expectation is often a quote generated from current on-chain pool state, expected fees and a selected route; the actual fill depends on the pool state when that transaction is included in a block. Official [Uniswap documentation](https://support.uniswap.org/hc/en-us/articles/8643794102669-Price-Impact-vs-Price-Slippage) separates slippage from price impact, and its [swap guide](https://developers.uniswap.org/docs/get-started/concepts/traders/swaps) describes user-side minimum-output and deadline checks.

### Quote versus fill: the simplest illustration

An interface quotes **1.0000 ETH** for a given input. Before execution, another transaction changes the pool state. Your trade settles for **0.9960 ETH**. Holding the quoted amount constant as the benchmark, the change is **−0.4% output slippage**: `(0.9960 / 1.0000 − 1) × 100`. If the transaction instead settles for **1.0030 ETH**, the same calculation gives **+0.3% positive slippage**, or price improvement. Not every order mechanism distributes positive improvement to the user in the same way; verify the execution terms, not just a generic DEX label.

This number does not tell you whether the *initial* 1 ETH quote was a good price versus the external market. An already-poor rate and a perfectly unchanged fill can produce **zero measured quote-to-fill slippage** but still be a bad deal.

### Positive, negative and zero slippage

**Negative slippage** is an execution worse than the displayed reference quote. **Positive slippage** is a better fill than quoted. **Zero slippage** means the filled amount equals the quoted reference within measurement precision, not that the trade was free. Quotes can contain trading fees and expected price impact, and the transaction may pay gas. A user buying a very illiquid token may experience nearly zero *additional* slippage after signing, yet pay a large execution premium already embedded in the quote.

### Why does slippage occur between signing and confirmation?

The user requests a quote from a specific market state, reviews it, authorizes the route and broadcasts or submits an order. Other swaps, liquidity changes, arbitrage, volatility, new blocks, reorderings and delays can change the state by the time a conventional on-chain swap executes. The amount displayed earlier is an estimate under a snapshot of conditions, not a timeless offer. A transaction deadline and an on-chain minimum-output check narrow the risks, but they cannot remove volatility or guarantee immediate execution.

### Does slippage happen on centralized exchanges too?

Yes. A centralized exchange market order can consume multiple price levels in an order book. The realized average execution price may differ from the last trade price, top-of-book quote or price seen before submission. This outcome depends on spread, depth and matching-engine timing. It is not automatically the same mechanism as an AMM reserve curve, but the practical lesson is the same: **compare filled quantity and realized average price with a clearly stated reference quote**.

## Slippage vs Price Impact vs Spread vs Fees: Which Number Matters?

The four terms describe **different layers of execution**. Use them together to reconstruct the complete transaction rather than treating them as interchangeable or simply summing percentages shown by unrelated interfaces.

| Term | What it measures | When it appears | Can it be limited? |
| --- | --- | --- | --- |
| **Reference price / mid-price** | A price before the trade, observed on a specified venue or external market | Quote comparison | Not controlled by a single swapper |
| **Price impact** | How the trade itself moves through liquidity and changes the average execution relative to the reference | Often **inside** the quoted output | Reduce size, find deeper liquidity or split routes; no universal guarantee |
| **Swap / LP fee** | Fee defined by the pool or routing mechanism | Normally **inside** a conventional DEX quote | Choose a different legitimate fee tier or route, subject to liquidity and trust |
| **Slippage** | Difference between the relevant quoted outcome and actual fill | Between quote and execution | Min-out, max-in, limit price or intent-order constraints can bound specified outcomes |
| **Bid-ask spread** | Difference between executable buying and selling prices | More visible in order books; can also emerge through AMM fee and pricing | Compare executable quotes on both sides |
| **Network gas** | Blockchain execution and data inclusion cost | Separate from output-token amount; often charged even if a direct trade reverts | Network choice, transaction size and fee management; not eliminated by tolerance |
| **Routing / protocol / hook fees** | Extra charge imposed by intermediary, hook, order mechanism or protocol | Varies by route and product | Inspect route details and compare effective outputs |
| **MEV extraction** | Value extracted through transaction inclusion, ordering or execution strategies | Possible around public on-chain swaps | Min-out constraints, route selection and certain protective execution systems can reduce exposure |

### Price impact is usually known at quote time, but slippage is not

Suppose 10,000 USDC could buy **4 ETH** at a mid-price of $2,500 with unlimited liquidity and no fees. The actual constant-product pool quote may be only **3.9486 ETH** after the pool's 0.3% fee and its own finite-reserve price impact. If that quote executes unchanged, additional quote-to-fill slippage is approximately **0%**, although the trader already receives roughly **1.28% fewer ETH than the idealized spot-price benchmark**. This comparison uses the number of output tokens as denominator; an inverse execution-price comparison gives a different percentage because percentages are not symmetric under inversion.

The [official Uniswap glossary](https://developers.uniswap.org/docs/get-started/concepts/glossary) and [Uniswap trading guide](https://developers.uniswap.org/docs/get-started/concepts/traders/swaps) both emphasize that price impact is inherent to trading through finite AMM liquidity. An actual interface may summarize components differently from our analytic labels, so **read the metric definition** before comparing percentages across platforms.

### Why double-counting ruins a swap-cost estimate

A user sees a *quoted* output after a pool fee and after the expected curve movement, then subtracts another displayed “price impact” percentage and a separate fee percentage from the same output. That understates the tokens they should expect, because those effects have already informed the quote. A clean model starts with an **independent reference price**, then computes the **actual quote** once, then applies only additional execution deterioration or other expenses that are genuinely excluded from the quote.

### Three useful prices to write down

1. **Independent reference price:** the external executable or fair-value benchmark, time-stamped, on an equivalent chain and asset.
2. **Quoted effective price:** input-token units divided by expected output-token units, or the inverse when appropriate; state whether fees are included.
3. **Executed effective price:** actual input divided by actual output from the receipt, plus any separately paid costs when comparing economic value.

If these three are not based on the same asset, network, time and units, the comparison is invalid. A bridged asset and a native asset with the same ticker may not be interchangeable at the exact same dollar value.

## How an AMM Creates Price Impact: Constant-Product Math

A conventional two-asset Uniswap v2-style automated market maker (AMM) adjusts reserves according to the **constant-product relation**, `x × y = k`, before considering fee accounting. Here `x` and `y` are the quantities of two assets in a pool; trades alter their ratio. A swapper moving a significant fraction of the reserves necessarily walks along the pricing curve rather than buying every unit at the initial marginal price. This principle is described in the [Uniswap v2 pricing documentation](https://developers.uniswap.org/docs/protocols/v2/concepts/pricing).

### A reproducible ETH/USDC calculation

Assume an invented v2-style pool containing **1,000,000 USDC and 400 ETH**. Its initial marginal price is therefore **2,500 USDC per ETH**. A user enters an exact-input trade to swap USDC for ETH. For demonstration, set a fixed **0.30% swap fee**. This is *not* a live Uniswap v4 fee and is not a current pool quote. Define `dx` as input USDC and let the effective amount used by the invariant be `dx_effective = dx × (1 − 0.003)`. Then the simplified fee-adjusted amount of ETH received is:

**`ETH_out = ETH_reserve × dx_effective / (USDC_reserve + dx_effective)`**.

The no-fee hypothetical spot benchmark is **`ETH_spot = dx / 2,500`**. The execution cost relative to this ideal spot benchmark, expressed as an **output shortfall**, is **`1 − ETH_out / ETH_spot`**. This includes both the simulated pool fee and the curve's finite-liquidity effect. It does *not* include time-dependent slippage, external routing costs or gas.

### The $100 / $1,000 / $10,000 Depth Stress Grid

| Input USDC | Ideal ETH at $2,500 | Quoted ETH after fee & curve | Effective USDC/ETH | Output shortfall vs spot | Dollar-equivalent shortfall |
| ---: | ---: | ---: | ---: | ---: | ---: |
| $100 | 0.040000 | 0.039876 | $2,507.77 | 0.310% | $0.31 |
| $1,000 | 0.400000 | 0.398403 | $2,510.02 | 0.399% | $3.99 |
| $10,000 | 4.000000 | 3.948632 | $2,532.52 | 1.284% | $128.42 |

When the input is only $100, the curve is barely affected. At $10,000, the input is **1% of the initial USDC reserve**, and the quote gives materially fewer ETH than the idealized spot conversion. This is the basic reason the same slippage setting can be acceptable for a small payment but inadequate protection for a large trade: the **quoted price impact** must be evaluated first, separately from the later **slippage tolerance**.

### Repeat the same order in a shallower pool

Now assume another invented pool at exactly the **same initial marginal price**, but with only **100,000 USDC and 40 ETH**. The fee is the same **0.30%**. This pool has one-tenth the depth of the previous example. It can still show ETH at an initial marginal price of $2,500, but a $10,000 trade moves much farther along its curve.

| Input USDC | Ideal ETH at $2,500 | Quoted ETH after fee & curve | Effective USDC/ETH | Output shortfall vs spot | Dollar-equivalent shortfall |
| ---: | ---: | ---: | ---: | ---: | ---: |
| $100 | 0.040000 | 0.039840 | $2,510.02 | 0.399% | $0.40 |
| $1,000 | 0.400000 | 0.394863 | $2,532.52 | 1.284% | $12.84 |
| $10,000 | 4.000000 | 3.626444 | $2,757.52 | 9.339% | $933.89 |

**The result is not a new slippage prediction.** Both tables represent the expected *quoted* execution at a fixed pool state and exclude time-dependent slippage entirely. Increasing the slider from 0.5% to 5% does not repair the shallow pool's already-poor quote. The economic remedy is to obtain deeper accessible liquidity, reduce trade size, choose a genuinely better route, or avoid the trade.

### Why these percentages vary across AMM designs

A v2 constant-product pool, a concentrated-liquidity v3 position and a customizable v4 hook may expose different liquidity amounts at successive price points. A dollar figure such as “$1 million TVL” is not enough to reconstruct a large swap without knowing **how much liquidity is active around the prices the swap traverses**. Some v4 hooks modify fee logic or custom accounting; multiple hops can change the route again. Our v2 formula is intentionally narrow in scope. Do **not** apply its result as a universal simulator for v3, v4, stable-swap curves or professional market-maker orders.

## How to Calculate Slippage Tolerance and Minimum Received

For a common **exact-input on-chain swap**, the trader specifies the amount to spend, receives a quote for the output and selects a maximum permitted deterioration. In simplified form, **`minimum_output = quoted_output × (1 − tolerance)`**, with tolerance written as a decimal. In production, token decimals and on-chain rounding matter; the router should enforce the correct integer token amounts. The minimum is an **execution floor**, not a promise that the trade will settle or that the external market price is fair.

Use the $10,000 USDC quote from our deep-pool example. The estimated output already includes the stipulated 0.30% fee and expected constant-product price impact:

| Extra tolerance | Quoted ETH (fee & price impact included) | Minimum output ETH | Maximum additional ETH shortfall | Reference-value shortfall |
| ---: | ---: | ---: | ---: | ---: |
| 0.1% | 3.948632 | 3.944684 | 0.003949 | $9.87 |
| 0.5% | 3.948632 | 3.928889 | 0.019743 | $49.36 |
| 1% | 3.948632 | 3.909146 | 0.039486 | $98.72 |
| 3% | 3.948632 | 3.830173 | 0.118459 | $296.15 |

The dollar shortfalls in this table value the **difference from the quote** at the hypothetical independent **$2,500/ETH** reference price. They are **not extra transaction charges** and do not predict the realized fill. A swap may execute at the full quoted amount, between the quote and its minimum, or revert if pool conditions deteriorate beyond the allowed output floor.

### What is a “good” slippage tolerance: 0.1%, 0.5% or 1%?

**There is no universally good setting.** The answer depends on the trade type, pool depth, price volatility, inclusion latency, network conditions, token behavior and available execution protections. A relatively tight **0.1%** can reject normal market noise in a volatile pool. **0.5%** may be a reasonable *example to examine*, not a blanket recommended number. A **1%** tolerance accepts twice the quoted-output deterioration of a 0.5% tolerance in a direct exact-input swap; it does not make an illiquid pair safer. The [Uniswap help page](https://support.uniswap.org/hc/en-us/articles/8643879653261-How-to-change-slippage-on-the-Uniswap-Web-app) documents interface-dependent auto slippage and the risks of setting a limit too low or too high. Its current auto-setting range is not an instruction to copy that range for every trade.

A meaningful decision starts by comparing **absolute worst-case tokens** and **equivalent value**. On a $25 trade, even a one-percent adverse move may be economically smaller than the network gas cost. On a six-figure trade, the same percentage can exceed thousands of dollars. A preset percentage without a trade-size context is incomplete.

### Exact-input versus exact-output swaps

For an **exact-input** trade (“spend exactly 10,000 USDC”), the common user protection is a **minimum output**: `amountOutMinimum`. For an **exact-output** trade (“receive exactly 4 ETH”), the protection is typically a **maximum input**: `amountInMaximum`. Under a simplified tolerance example of 0.5%, a target of 10,000 USDC expected input might allow up to **10,050 USDC**, if the interface uses a direct `expected_input × (1 + 0.005)` rule. However, **interfaces can calculate tolerance differently**; always inspect the exact max-in shown and enforced by the route. The [Uniswap swap guide](https://developers.uniswap.org/docs/get-started/concepts/traders/swaps) and [API quote reference](https://developers.uniswap.org/docs/api-reference/aggregator_quote) distinguish the two trade types.

### What happens if minimum output is not met?

For a conventional directly submitted on-chain swap with enforced min-out and no alternative execution path, the swap usually **reverts**, meaning the proposed asset exchange does not complete. But blockchain computation may still consume gas. A failed transaction should not be dismissed as “free protection”; it has a cost, and repeated aggressive retries can be uneconomic. An approval transaction may be separate from the swap and can have its own gas cost and security significance.

### Why deadlines matter as much as percentages

An on-chain transaction deadline prevents a stale swap from executing after a configured time, where supported and enforced. A stale quote can be dangerous even if the token pair is generally liquid. But setting a very short deadline may increase failed attempts during congestion, and a long one increases exposure to delayed execution. The [Uniswap v3 guide](https://developers.uniswap.org/docs/protocols/v3/guides/swapping/single-hop-swapping) covers deadlines, minimum output and optional price-limit parameters. The protection is integration-specific; check what the actual transaction is enforcing.

## The Terenval Quote-to-Receipt Ledger: An Original Swap Audit

Most swap interfaces display an attractive token output, a price impact figure, a fee line and a network fee estimate—but not every number necessarily shares a denominator or includes the same services. We propose the **Terenval Quote-to-Receipt Ledger**, an educational record that connects a quote to its finalized transaction.

| Ledger layer | Record exactly | Why it matters |
| --- | --- | --- |
| 1. Asset identity | Chain ID, token contract addresses, input/output decimals, native or bridged representation | Same ticker can represent different assets or networks |
| 2. Reference market | Independent time-stamped price, denomination and whether executable | Shows if the initial quote is already poor |
| 3. Quote | Input amount, estimated output, expected effective execution price, expiry or quote age | Defines the baseline before submission |
| 4. Route | Pools, fee tiers, intermediate tokens, optional hook/solver/aggregator, expected gas | Identifies how execution is expected to happen |
| 5. User guardrails | Minimum output or maximum input, deadline, limit-price constraints | Defines the maximum adverse fill the transaction itself permits |
| 6. Authorization | Approval or permit scope, spender, chain, expiration if applicable | Shows who can move tokens and under what conditions |
| 7. Receipt | Actual token deltas, status, block, gas, separate fees and asset changes | Establishes what really happened on-chain |
| 8. Outcome comparison | Output-versus-quote, output-versus-independent benchmark, total transaction cost | Separates quote quality from execution movement |

A good audit can show both **quote cost** and **execution drift** independently. The real user loss relative to an external benchmark is not necessarily the sum of named interface percentages because some metrics overlap. After an exchange, record actual wallet balance changes—not only a green “success” toast—and check whether the output token is transferable and genuine.

### Quote-to-receipt worked example

Suppose the 10,000 USDC deep-pool quote returns **3.9486 ETH**. Your exact-input contract protects a minimum **3.9289 ETH** with 0.5% tolerance. If the transaction settles for **3.9400 ETH**, the realized quote-to-fill movement is approximately **−0.22%**, while the shortfall from the ideal **4 ETH at $2,500** is about **1.50%**. These measure different comparisons. If the direct transaction also spends a hypothetical **$8 in gas**, its total cost versus the idealized reference is not the same as the 0.22% quote slippage alone.

**This is illustrative, not a claim about actual trade history.** If the wallet held ETH or USDC as a gas asset, or if a relayer paid the fee and recovered value elsewhere, the economic calculation should account for that correctly rather than adding fictitious gas charges to both parties.

### A compact framework for financial reporting

For a fixed-input quote evaluated in units of the purchased asset:

- **Quote deterioration:** `1 − quoted_output / ideal_reference_output`.
- **Execution drift:** `1 − realized_output / quoted_output`.
- **Total output shortfall:** `1 − realized_output / ideal_reference_output`.

These do not add exactly: the third is **`1 − (1 − quote_deterioration) × (1 − execution_drift)`** when all denominators refer to the same token and no other adjustments apply. That multiplicative relationship is an original reporting choice designed to avoid rounding and double-counting mistakes. Separately paid gas and service fees should be valued in the same unit before reporting a comprehensive net result.

## What Causes High Slippage on Decentralized Exchanges?

The most useful classification separates (a) an **unfavorable quote before signing** from (b) **deterioration while the trade is pending**. Both can leave a user with fewer tokens than an idealized reference conversion, but the remedies differ.

### Shallow pool liquidity

A pool with only 40 ETH and 100,000 USDC cannot absorb a 10,000 USDC order at anything close to its current marginal rate under a v2 constant-product curve. If no better liquidity is available, lowering a slippage tolerance will usually make the user reject or revert an unfavorable execution—it will not create liquidity. A practical remedy is to compare legitimate venues and routes, reduce size, use a suitable limit or intent order, or postpone the trade.

### Volatile token prices and stale quotes

A quote based on one block can be outdated by the next. A fast-moving token, an unexpected liquidation cascade or a new market announcement can change the acceptable fill within seconds. Re-request the quote before signing and avoid keeping stale approvals or transaction drafts open indefinitely. Freshness matters independently of the pool's total TVL.

### Congestion and transaction ordering

On a public blockchain, submitted swaps compete for inclusion. The eventual transaction ordering may differ from the order in which wallets displayed quotes, and conditions can change while transactions are pending. Paying more gas can influence inclusion probability under the relevant fee market, but it does **not** guarantee a particular price, make a transaction immune to MEV, or prove the chosen router is safe.

### Large order relative to active price-range liquidity

Concentrated-liquidity pools may hold substantial TVL while only a fraction is positioned along the price intervals crossed by a large swap. As an order moves through successive ticks, execution prices and fees can change. Users should compare executable outputs, not merely TVL or an advertised average fee tier.

### Token transfer taxes, rebasing and non-standard behavior

Some tokens deduct a fee when transferred or use unusual token accounting. A quoted swap may not fully model every token's taxes or transfer restrictions. Raising slippage to very high levels can expose the trader to worse fills without resolving a malicious or non-transferable asset. Carefully verify the token contract, chain and eligibility for the selected router before attempting a swap.

### Fee and hook changes

Uniswap v4 pools can employ dynamic LP fees and optional hooks that customize swap behavior. The fee seen when a quote is generated might depend on contract state or an attached hook, and cannot be assumed equal to a generic historical 0.3% figure. For risk analysis, review the actual [v4 dynamic fee documentation](https://developers.uniswap.org/docs/protocols/v4/concepts/dynamic-fees), the fee charged in the simulated transaction and any external dependencies. A low displayed slippage tolerance is not a substitute for understanding the pool code.

## MEV and Sandwich Attacks: How Execution Can Be Manipulated

**Maximal extractable value (MEV)** describes value obtainable through control over transaction inclusion, ordering or exclusion in block production. Not all MEV reflects theft or user harm; some arbitrage helps keep prices aligned. But **sandwich attacks** can directly worsen a user's DEX execution by exploiting a pending swap and its tolerance band. The [Ethereum.org MEV overview](https://ethereum.org/developers/docs/mev/) distinguishes useful arbitrage from harmful sandwiching and systemic network consequences.

### How a sandwich works, conceptually

1. A user's pending public swap reveals the direction, route and permitted worst-case result.
2. An attacker or searcher identifies a way to trade **before** it so the victim's route offers a worse execution price.
3. The victim's transaction executes inside its on-chain tolerance, if the manipulated terms still satisfy the minimum-output or maximum-input conditions.
4. The attacker may trade back **after** the victim, capturing some value created by the temporary price distortion, net of fees and other costs.

The ordering matters. **A normal adverse price move between quote and fill is not itself proof of a sandwich.** Establishing one requires transaction ordering, relevant pool state, counterparties and traceable transfers. An attack can also fail or be unprofitable; the presence of a suspected searcher does not establish realized attacker profit.

### A same-direction preceding trade changes the quote

Suppose our user expects the $10,000 USDC-to-ETH deep-pool quote, but another swap buying ETH for **20,000 USDC** enters the pool first. A reserve-model simulation predicts that the user's subsequent output will deteriorate because the ETH reserve is lower and the USDC reserve higher. That preceding trade could be a legitimate trader, an arbitrageur or part of an adversarial sequence; without observing a matching after-trade and other evidence, **we cannot call it a sandwich attack**.

| Scenario | User USDC input | User received ETH | Deviation from the original user quote |
| --- | ---: | ---: | ---: |
| User trades from original state | $10,000 | 3.948632 | 0.000% |
| Another $20,000 USDC buy executes first | $10,000 | 3.796260 | -3.859% |

The example highlights *ordering sensitivity*, not a measured amount stolen on a real blockchain. A protective minimum-output floor can reject the later execution if the deterioration exceeds tolerance, but users must still weigh gas lost to any revert, changes to the quote and other affected tokens.

### Does high slippage tolerance attract sandwich attacks?

It can make a visible exact-input swap easier to exploit, because a wider acceptable execution band gives an adversarial route more room to worsen the user's terms while still allowing success. However, profitability depends on pool depth, trade size, fees, builder participation and market conditions. **Not every successful trade with high tolerance is sandwiched**, and a tight tolerance cannot eliminate all forms of MEV or guarantee fill availability.

### Why “private transaction” is not an absolute promise

Private order routing can reduce public-mempool exposure, but actual protections depend on who receives the order, how that order is passed to block builders or solvers, fallbacks, trust assumptions, cancellation and how the transaction is eventually included. A route may use a private relay yet have other MEV or information-leakage risks. Check the provider's documented protection scope and status rather than assume all private RPCs deliver equivalent guarantees.

### Gas bidding is not a MEV defense by itself

A high priority fee may influence inclusion, but it does not by itself prevent a searcher from constructing a profitable ordering. Conversely, setting near-zero gas to avoid MEV can leave the transaction pending longer and expose the quote to more changes. The strategy should focus on **enforceable price bounds and execution design** rather than the misleading idea that paying more gas always avoids a sandwich.

## The Terenval MEV Exposure Gate: A Practical Review

The **MEV Exposure Gate** is an original editorial checklist, not a bot, protocol feature or automated guarantee. It asks whether a proposed swap is easy to observe, valuable to target and allowed to execute at a meaningfully worse rate.

| Gate | Evidence to inspect | Why it affects risk |
| --- | --- | --- |
| **Visibility** | Public pending transaction versus an authenticated private/intent route | Public visibility can make the swap easier to target |
| **Value at risk** | Trade notional and tokens potentially lost at minimum output | Larger allowed deterioration can create more extractable value |
| **Market depth** | Size relative to active pool liquidity | Low depth makes price more sensitive to preceding trades |
| **Allowed movement** | Min-out, max-in, deadlines and any limit price | Wide limits permit more harmful executions |
| **Route architecture** | Direct AMM, multi-hop, pool hook, aggregator, RFQ, solver or auction | Different designs change information exposure and trust assumptions |
| **Protection verification** | Specific documented routing guarantees and result receipts | Avoid relying on an unspecified “anti-MEV” badge |

Our qualitative rule is to **increase scrutiny** when a large trade is publicly visible, route liquidity is thin and the permitted minimum output is substantially worse than the reference quote. A private route can lower some risks without curing a bad quote, a malicious token or an incorrect approval. This gate deliberately has no arbitrary numeric “safe” threshold.

## Private Routing, Intent Orders, UniswapX and CoW: What Changes?

Conventional on-chain AMM swaps are not the only execution format. Some systems collect **orders or intents**, solicit competing fills or settle transactions using auction designs. Their protection is often expressed as **signed limit conditions, minimum outputs and solver competition**, not the same single slippage slider used for an immediate AMM swap.

### UniswapX: signed order terms rather than the ordinary slider

Uniswap's official [UniswapX support documentation](https://support.uniswap.org/hc/en-us/articles/20071671202189-Can-I-change-slippage-when-using-UniswapX) says the ordinary interface slippage setting **does not apply** to swaps using UniswapX. Users can instead inspect the **minimum amount of tokens** the order may receive. Its [protocol description](https://support.uniswap.org/hc/en-us/articles/17515415311501-What-is-UniswapX) describes auction-based liquidity sourcing and MEV-related protections. It is inaccurate to tell all users “set your UniswapX tolerance to 0.5%,” because that setting is not the relevant order control.

UniswapX support, available order types, settlement terms, network cost handling and quote freshness depend on the current interface and chain. Read exactly what is signed and who may fill the order. The presence of an auction does not certify a token or eliminate every possible execution risk.

### CoW Protocol: batch competition and coincidence of wants

[CoW Protocol's documentation](https://docs.cow.fi/cow-protocol/concepts/introduction/fair-combinatorial-auction) describes batches of signed intents and competing solvers. Some orders can match directly through a **coincidence of wants** without taking liquidity from an AMM, while others require on-chain liquidity sourcing. This design can reduce certain forms of transaction-order exploitation. The [solver rules](https://docs.cow.fi/cow-protocol/concepts/introduction/solvers) enforce signed conditions and settle through protocol contracts.

The trade-offs include auction timing, order expiration, solver behavior, allowance permissions and liquidity availability. A specific order may remain unfilled or be settled through an AMM. Compare **actual executable minimum terms** and total costs rather than treating “MEV-protected” as a universal guarantee of the best price.

### RFQ, limit orders and solver quotes

A request-for-quote (RFQ) can seek a price from a maker for a specified amount and validity period. A limit order can refuse trades outside a user-defined rate, at the cost of non-execution. Solver systems can compete for favorable routes or quote surplus. These mechanics change who bears market-movement risk and who can improve the fill, but require attention to **expiry, partial fills, signature scope, allowance and cancellation**. A better displayed rate is only useful if it can be executed securely and settled as specified.

### A neutral comparison

| Mechanism | Typical user control | Potential benefit | Main caveat |
| --- | --- | --- | --- |
| Direct AMM exact-input swap | Min output + deadline | Immediate on-chain execution if conditions permit | Public exposure, pool impact, gas on revert |
| Direct AMM exact-output swap | Max input + deadline | Fixed desired output if executable | Input cost can rise up to cap; may revert |
| Limit order | Minimum acceptable execution price | Rejects worse fills | May not execute when market moves away |
| Auction/intent-based order | Signed order constraints and solver competition | Alternative liquidity and some MEV mitigation | Different timing, permissions, fill conditions |
| Private transaction route | Route-specific private submission and on-chain constraints | Reduces some public observation | Relay/builder trust, fallbacks and non-guaranteed inclusion |

**Terenval's position:** Evaluate protections by their **enforceable settlement conditions** and observed execution, not their branding. Claims such as “slippage-free,” “no MEV ever” or “guaranteed best price” require clearly documented scope and limitations before readers should rely on them.

## The Terenval Slippage Exposure Budget

Instead of picking a tolerance first, calculate the **absolute cost of what you are allowing**. Our editorial **Slippage Exposure Budget (SEB)** expresses the extra token amount at stake from the quote to the worst permitted outcome for a conventional exact-input swap:

**`SEB_tokens = quoted_output − minimum_output`**; for a simplified percentage limit, **`SEB_tokens = quoted_output × tolerance`**.

This is *not* the total cost of the trade. It excludes price impact already in the quote, quoted swap fees, network gas and hidden or dynamic service fees. It is a maximum **additional output deterioration relative to the quoted amount** that the specific on-chain minimum-output constraint would allow, assuming it is actually enforced and no unusual token behavior interferes.

### Converting the exposure budget into reference value

Suppose you quote 3.9486 ETH and accept 0.5% further adverse output movement. That permits a loss of roughly 0.01974 ETH relative to quote; at our example benchmark price of $2,500, that's about **$49.36 of reference value**. The user should judge that amount against the purpose and size of the trade. On a 100 USDC trade in the same deep pool, a 0.5% output tolerance represents a much smaller absolute amount; the same percentage has a different economic meaning.

### Tolerance is not the same as a capital risk budget

A token can later collapse in price after a perfectly executed swap, be frozen by its issuer or have malicious contract logic. No slippage tolerance caps those **post-trade** losses. Similarly, if a bridge or aggregator charges an additional fee that was not reflected in the quote, the SEB should not be misrepresented as a global all-in loss limit. Use it as one line item inside the larger Quote-to-Receipt Ledger.

### A decision threshold without a magic percentage

Before authorizing, ask: **Would I voluntarily pay the maximum additional number of tokens this transaction permits, even if no attack were present?** If not, revise the trade size, venue, timing, route or order format. Repeatedly increasing tolerance to force success is not a strategy for improving liquidity or avoiding a sandwich; it just changes the permitted execution terms.

## The Terenval Route Decision Framework: Choose the Outcome, Not the Brand

Another original editorial model, the **Terenval Route Decision Framework**, compares candidate routes by **user-receivable value under the same market benchmark**. In many cases a quote with a nominally lower fee delivers fewer tokens because its price impact is worse, while a route with a higher fee but deeper liquidity can win. A route using a single pool may be cheaper in gas than a multi-hop route even when its swap-price quote is marginally worse.

### First: compare equal input, output and asset identity

Start every comparison with the same input token contract, input amount, destination token contract, chain and quote timestamp. “ETH” on one chain and a bridged wrapper named ETH on another may not represent the same redeemable asset. Likewise, receiving USDC versus a different stablecoin changes the comparison. If the asset and final network differ, include bridge, conversion and settlement assumptions explicitly.

### Second: compare output after swap fees, before gas

For the 10,000 USDC example, consider the one-pool deep route versus two equally deep independent pools that share the same starting marginal price and fee. Splitting the order can reduce marginal movement if both routes truly provide incremental liquidity. In our deliberately idealized model, each independent pool begins with **1,000,000 USDC and 400 ETH** and charges **0.30%**. One route uses a single pool for all 10,000 USDC; the split route uses **5,000 USDC in each of the two pools**. The comparison excludes extra execution costs and assumes no changes between quoting and filling.

| Hypothetical route | Output ETH before gas | Gain over single-pool route | Equivalent value at $2,500/ETH |
| --- | ---: | ---: | ---: |
| Single deep pool, 10,000 USDC | 3.948632 | — | $9,871.58 |
| Split into two independent equally deep pools, 5,000 each | 3.968218 | +0.019586 ETH | $9,920.55 |
| Split-route advantage before extra gas/service costs | — | +0.019586 ETH | +$48.97 |

**The split is not universally superior.** If two labeled pools share the same underlying reserves or the route requires an expensive additional hop, there may be little or no executable advantage. A routing engine's displayed price can also become stale. This is a controlled teaching example of independent depth rather than a recommendation to manually split transactions.

### Third: add real gas, bridge and service costs

Convert network gas to the same reference unit at the relevant token prices and confirm who actually pays it. A multi-hop route might provide an extra $12 of output but require $20 more gas; that can make it inferior on a net basis. A solver-funded gas transaction may recover its cost through quoted terms or fees rather than a visible gas debit. Compare **net tokens and settlement conditions**, not “gasless” labels alone.

### Fourth: inspect risk before choosing the top quote

A suspicious pool can advertise a spectacular exchange rate while the received token is non-transferable. An unfamiliar v4 hook may expose custom execution logic. A bridge may offer a better displayed rate but introduce additional trust or withdrawal delays. The best route is not merely the maximum reported output: it must also be **valid, executable, correctly authorized and acceptable under the user's risk policy**.

## How Slippage, Fees and Gas Change Across Ethereum and Layer 2

A user moving from Ethereum mainnet to an L2 may reduce gas costs while encountering different liquidity depth, bridge costs, sequencer practices and trade routes. **Lower network gas and lower slippage are separate propositions.** A thin L2 pool may give a materially worse ETH/USDC quote than a deep Ethereum-mainnet pool, despite costing less to execute. Conversely, an active L2 pool can be highly competitive.

### Why “L2 always has cheaper swaps” is misleading

A fair all-in comparison includes the cost of moving assets to the target chain when relevant; local token representation; pool price impact; fees; the swap's gas; possible sequencer or settlement delays; and the final destination of the tokens. If the user ultimately needs assets back on another chain, bridge exit constraints and related costs matter. A displayed “$0.02 network fee” does not establish that a $10,000 exchange is economical if the quote gives $150 less output than the alternative.

### Fragmented liquidity across networks

Liquidity can be scattered across different chains, fee tiers, pool designs and token wrappers. An aggregator may help source routes, but no single aggregator can guarantee best execution across all venues at all times. The useful benchmark is the **actual amount of the intended final asset delivered**, after any extra bridging and service steps, under a clearly defined time and risk horizon.

### MEV on L2 is a different operational environment, not nonexistent

Sequencing design and transaction visibility differ across rollups and chains. This may alter how certain adversarial transaction-ordering strategies operate but does not remove execution manipulation, frontrunning risk or adverse price changes by definition. Assess the actual chain and route architecture instead of extrapolating every Ethereum-mainnet MEV rule unchanged.

## Stablecoin Swaps, Depegs and Hidden Premiums

An apparently low-slippage USDC-to-USDT trade can still carry substantial economic risk if the reference tokens no longer have equal market value or redemption quality. Consider a hypothetically depegged stablecoin worth **$0.95** while another remains near **$1.00**. Exchanging 1,000 units of the weaker token for 950 units of the stronger token may accurately reflect external value. The quote-to-fill slippage could be zero, even though the user receives only 950 units for 1,000 input units. That value difference is **not automatically 5% slippage**, because the two starting assets are no longer equal in price.

### Asset-unit output versus dollar purchasing power

Always distinguish the *number of received tokens* from their true market or redemption value. A million units of a malicious meme token with no viable sell-side liquidity may be less valuable than one genuine USDC. A UI reporting a favorable token quantity is not enough to prove good execution or even a usable asset.

### Concentrated stablecoin liquidity and out-of-range prices

Stable-pair pools may be deep close to their intended ratio and much less forgiving beyond certain intervals, especially when liquidity is concentrated in narrow bands. An abrupt depeg can move an exchange through ranges with limited active liquidity. Check both the quoted route and the value of the *actual asset being received*, not merely a historical stablecoin pool TVL.

### Fee-on-transfer tokens masquerading as slippage

A token that deducts units on transfer may make actual received balances differ from an interface's naive quote even without ordinary price movement. A failed swap or unexpectedly high tolerance is not sufficient evidence of whether this is a benign tax token, a misconfigured router or an intentionally abusive contract. Investigate the token contract and do not solve unexplained losses by repeatedly raising the slippage slider.

## How to Avoid Costly Crypto Swaps: Before, During and After

There is no universal method that guarantees the lowest price, no slippage, no failed transaction and no MEV loss. A **process** can materially improve decision quality while acknowledging the limits of the chosen route and wallet.

### Before requesting a swap

1. **Verify the token contract addresses and chain** from reliable independent sources. Token symbols and logos are not unique identifiers.
2. **Decide whether you care about exact input or exact output.** It determines which user-side bound matters.
3. **Obtain an independent price benchmark** for the same underlying assets and network at a matching time.
4. **Check pool and active-range depth**, not only dashboard TVL or nominal fee tier.
5. **Compare multiple reliable routes when economically justified**, including net output after real costs.
6. **Inspect any router, hook, vault or bridge dependency** rather than approving an unexplained spender.
7. **Record the expected output and worst permitted output** in token units before signing.

### While reviewing transaction authorization

1. Read the **input amount**, **minimum output or maximum input**, network, recipient, spender and any deadline.
2. Distinguish a simple wallet connection from an **ERC-20 approval** or signed permit authorization. The latter may grant token spending rights.
3. Reject requests to reveal the recovery phrase or private key. A DEX swap does not require handing those credentials to a website.
4. If price conditions have moved, **refresh the quote** rather than assuming the earlier price still applies.
5. Treat unusually high requested tolerance as a **reason to investigate liquidity, token mechanics or route design**, not automatically a reason to raise it further.

### After submission or settlement

1. Inspect the receipt to confirm **success or revert** and the actual token deltas.
2. Record the gas paid, including gas for any failed attempt.
3. Compare the final output with both the quote and the external market benchmark, using consistent units.
4. If a harmful sandwich is suspected, examine surrounding transactions and pool reserves before attributing motive or asserting a measured loss.
5. Review remaining token approvals and revoke unwanted permissions through a **verified** wallet or block-explorer interface where supported.

The Terenval educational guide on [checking and revoking token approvals](https://wallet.terenval.com/answers/how-to-check-and-revoke-token-approvals-in-a-crypto-wallet/) explains how authorization exposure differs from an adverse trade quote. Our separate guide to [crypto address poisoning](https://wallet.terenval.com/answers/what-is-crypto-address-poisoning-and-how-to-avoid-it/) covers a different recipient-selection scam that must not be confused with slippage or MEV.

## Terenval Wallet: Where Self-Custody Helps—and What It Cannot Guarantee

A self-custody wallet's central value is **control of signing authority for supported networks and accounts**, not a universal promise of favorable prices. Users can use a wallet as an authorization boundary: verify who requests the signature, which asset and chain are involved and whether the transaction's visible terms match the intended action. That awareness matters for swaps as well as DeFi lending and liquidity provision.

**Terenval Wallet** is a self-custody wallet in the Terenval ecosystem. In this article, it is mentioned natively as part of the broader practice of verifying signing requests and supported assets. **We do not claim that Terenval Wallet currently provides a built-in DEX aggregator, slippage control interface, Uniswap/UniswapX/CoW integration, private MEV routing, real-time price-impact warnings or automatic anti-sandwich protection.** Those are distinct product capabilities that must be individually verified in the actual software before making claims or connecting a third-party service. An attractive screenshot or promotional video is not proof of a deployed protection mechanism.

### Self-custody does not make a contract safe

A user can hold keys independently while signing a permission that lets a malicious spender transfer tokens, or while approving a transaction that sends funds into an unsafe pool or hook. The underlying wallet does not control the external contract's economic decisions. Review the actual spender, allowance amount and transaction effects, and avoid exposing seed phrases to supposedly mandatory “swap verification” portals.

### What a Terenval user should check before a hypothetical external DEX interaction

Confirm the relevant **coin, network and token standard are actually supported**; inspect the approval or permit's spender and scope; compare the quoted receive amount with an external source; and verify the on-chain minimum or maximum execution condition if the selected integration exposes it. For general cross-network mistakes, see our [guide to choosing the correct network for a withdrawal](https://wallet.terenval.com/answers/how-to-choose-the-correct-network-when-withdrawing-crypto-from-an-exchange/). The same discipline helps users of other self-custody wallets.

### A note on trust and transparency

Terenval's native presence in this guide is **educational and contextual**. A neutral article should not imply a wallet has exclusive pricing intelligence, direct protocol relationships, insurance or product defenses when those functions have not been independently established. Our editorial position is that users should demand verifiable transaction terms from **any** wallet and DEX, including ours.

## Seven Practical Scenarios: What a Trader Should Actually Do

The following are illustrative scenarios, not observed Terenval trades, live quotes or financial recommendations.

### Scenario 1: A $100 swap in deep ETH/USDC liquidity

In our 1,000,000-USDC reserve model, the $100 trade has almost negligible *curve* impact compared with $10,000. The pool fee is a greater proportional component than the curve movement. For such a small trade, gas could dominate the economics even if quote-to-fill slippage is near zero. Focus on realistic net value after network costs, not obsessively optimizing a few basis points while paying a disproportionate execution fee.

### Scenario 2: A $10,000 trade in a shallow new-token pool

A token pair shows the same initial marginal price as a deep pool but has one-tenth the reserves. The user gets a much worse quoted result due to price impact. Raising tolerance from 0.5% to 5% does not improve that initial quote; it simply authorizes more potential further deterioration. Find independent liquidity, reduce trade size or decline the swap.

### Scenario 3: A major market move between quote and block

The UI shows an attractive estimate. The user reviews and waits, and the price moves against the transaction. A min-out protects a defined floor, not the original market condition. Depending on the movement, the swap can settle within tolerance, revert while consuming gas or remain pending until expiry. Refreshing and re-evaluating the price before signing is better than treating yesterday's quote as a price commitment.

### Scenario 4: A low-fee pool is worse than a higher-fee pool

A 0.05% pool has shallow active liquidity; a 0.30% pool has substantial depth. For a large trade, the deeper pool may deliver more final tokens despite charging a higher nominal fee. Compare quoted **amount received**, not just fee-tier labels. If the shallow pool is manipulated or disconnected from external prices, a low headline fee can be a costly trap.

### Scenario 5: A high-tolerance public transaction attracts adversarial attention

A large exact-input swap is visible and permits a generous negative output move. An adversarial trader may be able to worsen the execution and profit from surrounding trades. Tightening a well-informed minimum-output floor limits the fill deterioration tolerated by the contract; a reputable private or auction route may reduce certain visibility risks. Neither option guarantees a fill or proves any particular transaction was sandwiched.

### Scenario 6: A “gasless” auction quote appears cheaper

An auction-based route advertises that the user does not pay a visible network fee, while another route shows gas and pool fees separately. The economically fair comparison is the **same final token delivered at the same settlement constraints**, including any fees embedded in the order. The solver-funded route may win, lose or fail to fill. “No gas charged directly to the user” is not proof of zero total economic cost.

### Scenario 7: An unknown token reports surprisingly good execution

The user sees a very large amount of a newly listed token at seemingly low slippage. The token might have a transfer tax, confiscatory control, extreme sell-side illiquidity or a counterfeit contract address. The swap can settle exactly according to the quote while leaving the user holding a token that cannot be sold. No slippage setting substitutes for verifying the token's identity, market and contract risks.

## Common Mistakes That Make Slippage Worse

- **Treating “0.5%” as a guaranteed fee.** It is usually a maximum permitted deviation relative to a quote, not a charge assessed in every successful swap.
- **Adding price impact twice.** The estimated output may already reflect the AMM's pricing curve.
- **Using the last traded price instead of executable liquidity.** The next $50,000 may not transact near the last $10 trade's price.
- **Ignoring which token the percentage is measured in.** Output shortfall and inverse execution-price premium are not numerically identical.
- **Assuming stablecoin means $1 forever.** A depeg changes the fair-value benchmark; it is not automatically “slippage.”
- **Increasing tolerance until an illiquid swap succeeds.** This may permit an unacceptable fill and raise susceptibility to adverse ordering.
- **Assuming every private RPC is MEV-proof.** Protection scopes, fallbacks and relay trust differ.
- **Confusing a reverted trade with free cancellation.** On-chain gas is often consumed even if token exchange fails.
- **Ignoring approval scope.** An unlimited spender authorization is not justified by a small expected trade.
- **Using an old quote after changing token, route or chain.** Compare identical trades and refresh state.
- **Believing APR or fee tier predicts user execution quality.** LP earnings and trader execution are different metrics.
- **Assuming the successful transaction cannot be harmful.** A correct contract may execute exactly the very poor price the user allowed.

## Why It Matters for Bitcoin, Ethereum, L2s, DeFi and Wallets

**Bitcoin (BTC):** Bitcoin network transfers are not inherently AMM swaps. Bitcoin traded on an order book can experience order-book slippage; wrapped BTC on EVM chains can experience AMM price impact, MEV and wrapper/bridge risks. These are different execution contexts, even if the ticker looks familiar.

**Ethereum (ETH):** Ethereum-native swaps may expose orders in a visible pending-transaction environment and require network gas. The user should inspect actual on-chain min-out/max-in constraints, price impact and token approvals. ETH market volatility alone is not proof of MEV; price changes and sandwich sequences must be separated in an audit.

**Layer 2:** Rollups and other L2s can offer lower per-transaction gas but differ in available liquidity, network mechanics, withdrawal options and token representations. The winning route is the one with acceptable **net final assets and risk**, not necessarily the lowest gas label.

**DeFi lending:** Collateral swaps and liquidations can require precise execution. Excessive slippage may magnify collateral-management losses or cause a protective transaction to revert during volatile conditions. For the related but different problem of borrowing and liquidation, see [How DeFi Lending Works](https://wallet.terenval.com/answers/how-does-defi-lending-work/).

**Liquidity providers:** Swappers' price impact, fee revenues and MEV interact with LP profitability. That is different from an LP's own **impermanent loss relative to holding original token amounts**; do not collapse them into the same metric. Our [impermanent-loss guide](https://wallet.terenval.com/answers/what-is-impermanent-loss-in-defi/) explains the holder-versus-LP comparison.

**Wallets:** A wallet protects or exposes the signing process depending on actual design and user behavior. It does not create deep pools, validate every token's economic integrity or make a signed market order immune to poor execution. Self-custody and swap quality are separate responsibilities.

## Frequently Asked Questions (FAQ)

### 1. What is slippage in cryptocurrency?

It is the difference between the result a trader expected from a quote or reference and the result actually executed. On a conventional DEX swap, this often reflects pool-price changes between the quote and transaction inclusion. Define the comparison in token or execution-price units before assigning a percentage.

### 2. Is slippage the same as price impact?

No. **Price impact** is the execution effect caused by your own trade relative to available liquidity; **slippage** compares a quoted outcome with the eventual fill. Some products describe total execution deviation using broader terminology, so check their definitions before comparing interfaces.

### 3. Does 1% slippage mean I lose 1% on every swap?

No. In a direct exact-input swap, it typically permits up to roughly 1% less output than the quoted amount before the trade is rejected. The actual fill may equal the quote, improve on it, worsen within the allowed band or fail to execute.

### 4. What is the best slippage tolerance for Uniswap?

There is no one-size-fits-all value. Evaluate the token's volatility, active liquidity, trade size, quote age, network conditions and independent reference rate. Inspect the actual minimum received, not just 0.1%, 0.5% or 1% labels; UniswapX orders use different controls.

### 5. Is 0.5% slippage always safe?

No. It does not remove expected price impact already in the quote, token taxes, malicious contracts, MEV risk or gas paid on failure. It may also be too tight for some volatile trades, causing reverts; the correct parameter depends on execution context.

### 6. What happens if I set slippage too low?

A conventional swap can fail or revert if execution terms cross the enforced minimum-output or maximum-input bound. Network computation can still cost gas. Repeated failures do not by themselves prove the token is legitimate or the pool safe.

### 7. What happens if I set slippage too high?

The transaction may be permitted to settle at a much worse price than the original quote. A generous band can also increase opportunity for harmful order manipulation. It cannot compensate for shallow liquidity or a poor reference quote.

### 8. How do I calculate slippage percentage?

For output-token comparison, use **`(received_output / quoted_output − 1) × 100%`**. A negative number indicates fewer output tokens than quoted. State your convention: some platforms report “positive slippage” differently or use inverse execution-price differences.

### 9. What is price impact on a DEX?

It is the deviation produced when a swap moves through finite available liquidity, compared with a pre-swap price benchmark. It tends to increase as the trade becomes large relative to accessible depth and can depend strongly on the pool's curve and active ticks.

### 10. Does slippage tolerance cover price impact?

For many direct AMM interfaces, the quoted output already reflects expected price impact; tolerance limits additional adverse movement around that quote. Do not assume every router uses identical definitions. Always inspect the enforced minimum or maximum tokens.

### 11. Why do large swaps have worse execution prices?

A large order consumes a greater fraction of available liquidity and moves farther through pool or order-book depth. Splitting across genuinely independent liquidity sources can sometimes improve the result, but extra fees, gas and execution delays can offset the gain.

### 12. Does adding liquidity reduce slippage?

More **accessible liquidity at the relevant prices** can reduce expected impact for a given trade, all else equal. Merely advertising a high total value locked does not guarantee the liquidity is active across the exact path used by the swap.

### 13. Is positive slippage possible?

Yes. The actual executed output can be greater than the quote if market conditions improve or a venue supplies price improvement. Whether a trader receives that surplus depends on the order mechanism and its signed settlement rules.

### 14. Can a crypto swap fail and still cost gas?

Yes. An on-chain transaction can consume gas even if an enforced min-out, deadline or another condition causes the swap to revert. Separately signed approval transactions may incur additional costs. Some order protocols have different user gas arrangements.

### 15. What is minimum received?

The minimum output-token amount permitted for an exact-input swap, when the transaction route enforces that bound. If the pool would produce less, a conventional protected transaction reverts instead of accepting the worse fill.

### 16. What is maximum input in a crypto swap?

For an exact-output order, it is the most input tokens the route is allowed to spend to obtain the requested output, under the specified contract or signed order conditions. Confirm it in actual token units.

### 17. What is a sandwich attack?

It is a harmful transaction-ordering strategy in which an attacker places a trade before a victim's swap to worsen its execution and another trade afterward to capture value. A bad fill alone is insufficient to prove such an attack occurred.

### 18. Can setting low slippage stop MEV?

A properly enforced low min-out tolerance limits the deterioration a successful swap can accept, but it does not prevent all ordering strategies, transaction failures, information leakage or gas costs. Effective mitigation may involve route design and privacy, with their own trust assumptions.

### 19. Are private RPCs completely MEV-proof?

No. Private routing reduces certain public visibility risks, but protection depends on relay/builders, fallback behavior and actual inclusion. Look for specific documented constraints and keep ordinary min-out/max-in safeguards where available.

### 20. Does UniswapX use the normal slippage slider?

No. Official Uniswap documentation says the ordinary slippage setting does not apply to UniswapX orders. Users should inspect the minimum-output and signed order terms of that auction-based execution path.

### 21. How is CoW Protocol different from a direct AMM trade?

CoW Protocol uses competing solvers and batch auctions to satisfy signed trading intents. Opposing orders can sometimes match directly; otherwise solvers can use external liquidity. Execution timing, permissions and price limits differ from a direct AMM router call.

### 22. Is a lower DEX fee always better?

No. A low-fee pool can be shallow enough to deliver worse net output than a deeper pool with a higher fee. Compare the exact quoted asset amount and all separate network/service costs using a consistent reference.

### 23. Is a low-gas Layer 2 swap automatically cheaper?

No. Lower gas can improve economics, but pool depth, token representation, bridge costs, fee tiers and route quality may dominate, particularly for larger swaps. Compare net delivered value on the intended final chain.

### 24. Is slippage the same as a stablecoin depeg?

No. A depeg changes the market value of an asset. A trade can have zero quote-to-fill slippage while exchanging a depegged token at a fair but unfavorable dollar rate. Evaluate redemption and asset-value risks separately.

### 25. Can a swap show 0% slippage and still be a bad deal?

Yes. The quote itself may embed large price impact, fees or an unfair exchange rate. Zero additional deterioration after signing does not establish a competitive initial quote.

### 26. Should I split a large swap into smaller trades?

Sometimes, but not automatically. Splitting can reduce impact across independent depth and enable staged execution, yet more gas, time exposure and potential MEV can make the result worse. Compare realistic executable routes and total costs.

### 27. What is the difference between slippage and spread?

Slippage is execution movement relative to a quote; the bid-ask spread is the difference between buying and selling prices at a point in time. Both can affect trade economics, but they are different measurements and may overlap in certain interface summaries.

### 28. Can token taxes cause apparent slippage?

Yes. Transfer taxes or other non-standard token behavior may change the received amount independently of normal AMM market movement. Verify the token contract and whether the router models this behavior; raising tolerance is not proof the trade is safe.

### 29. Does Terenval Wallet have built-in MEV protection or slippage settings?

This guide does **not** claim that Terenval Wallet implements a built-in DEX swap, Uniswap integration, MEV-resistant private routing or slippage controls. Terenval Wallet is mentioned in the context of self-custody and checking transaction authorization. Verify supported networks and features in the actual application.

### 30. What should I check immediately before confirming a DEX swap?

Verify the **correct token contracts and chain**, fresh quote, independent market benchmark, route and fee layers, expected output, **minimum received or maximum input**, deadline where relevant, network gas, contract spender and any permit scope. Decline the trade if the worst acceptable result is unacceptable—even if the interface reports the transaction can succeed.

## Related Terenval Reading

The [Terenval guide to impermanent loss in DeFi](https://wallet.terenval.com/answers/what-is-impermanent-loss-in-defi/) explains the distinct LP-side exposure created by holding a constantly rebalanced pool position. It complements this article because a swapper's **price impact and fee** become part of the liquidity-provider economics, but the two perspectives require different benchmarks.

Our [DeFi lending guide](https://wallet.terenval.com/answers/how-does-defi-lending-work/) explains collateral, loan health factors and liquidation—another workflow where forced swaps during volatile markets can amplify costs. If you need to understand transaction authorization before using third-party DeFi applications, see [how to check and revoke token approvals](https://wallet.terenval.com/answers/how-to-check-and-revoke-token-approvals-in-a-crypto-wallet/). For anti-fraud and transfer safety, use the guides to [crypto address poisoning](https://wallet.terenval.com/answers/what-is-crypto-address-poisoning-and-how-to-avoid-it/) and [choosing the correct blockchain network](https://wallet.terenval.com/answers/how-to-choose-the-correct-network-when-withdrawing-crypto-from-an-exchange/).

For foundational account safety, Terenval also covers [using the same crypto wallet in two apps](https://wallet.terenval.com/answers/can-you-use-the-same-crypto-wallet-in-two-apps/) and [seed phrase versus private key versus wallet password](https://wallet.terenval.com/answers/seed-phrase-vs-private-key-vs-wallet-password/). Those topics are separate from swap pricing; a secure seed phrase cannot repair an economically bad trade.

## Sources and Research Notes

The article favors **primary protocol and product documentation** for definitions, contract protections and execution architectures. The original calculation tables are independently derived from *declared hypothetical reserves*, not extracted from live pools. We distinguish official product claims about mitigation from independent proof of a specific result, and we do not treat documentation for one version or order type as universal.

1. [Uniswap Support — Price Impact vs Price Slippage](https://support.uniswap.org/hc/en-us/articles/8643794102669-Price-Impact-vs-Price-Slippage). Core definitions used to distinguish quote impact from execution drift.
2. [Uniswap Support — What Is Price Impact?](https://support.uniswap.org/hc/en-us/articles/40074715860365-What-is-price-impact). Liquidity depth and trade-size effects.
3. [Uniswap Developers — Understanding Swaps](https://developers.uniswap.org/docs/get-started/concepts/traders/swaps). Execution model, user-safety bounds and version-specific architecture.
4. [Uniswap Support — Changing Slippage Settings](https://support.uniswap.org/hc/en-us/articles/8643879653261-How-to-change-slippage-on-the-Uniswap-Web-app). Auto versus custom slippage and trade-revert behavior.
5. [Uniswap Support — Swap Screen Details](https://support.uniswap.org/hc/en-us/articles/39862756339341-Uniswap-Web-App-The-Swap-Screen). Quote, price impact, network cost and minimum-received fields.
6. [Uniswap Developers — v2 Pricing](https://developers.uniswap.org/docs/protocols/v2/concepts/pricing). AMM curve math and risks when the external reference price is ignored.
7. [Uniswap Developers — v2 SDK Trading Guide](https://developers.uniswap.org/docs/sdks/v2/guides/swapping). Minimum output, tolerance and deadline mechanics.
8. [Uniswap Developers — v3 Single Swaps](https://developers.uniswap.org/docs/protocols/v3/guides/swapping/single-hop-swapping). Price limits, deadline and production safety notes.
9. [Uniswap Developers — v3 Multi-Hop Swaps](https://developers.uniswap.org/docs/protocols/v3/guides/swapping/multi-hop-swapping). Routes, pool fees and multi-hop safety.
10. [Uniswap Developers — Fee Structures](https://developers.uniswap.org/docs/get-started/concepts/fees). LP fees, protocol fees and version-specific accounting.
11. [Uniswap Developers — v4 Hooks](https://developers.uniswap.org/docs/get-started/concepts/hooks). Optional custom pool logic and external contract behavior.
12. [Uniswap Developers — v4 Dynamic Fees](https://developers.uniswap.org/docs/protocols/v4/concepts/dynamic-fees). How pool fee rates can vary by hook and market conditions.
13. [Uniswap Developers — Protocol Glossary](https://developers.uniswap.org/docs/get-started/concepts/glossary). Definitions of price impact, slippage, routes and liquidity.
14. [Ethereum.org — Maximal Extractable Value](https://ethereum.org/developers/docs/mev/). MEV overview, useful arbitrage and harmful sandwiching.
15. [Uniswap Support — UniswapX Slippage Settings](https://support.uniswap.org/hc/en-us/articles/20071671202189-Can-I-change-slippage-when-using-UniswapX). Why the normal interface slider does not apply to UniswapX orders.
16. [Uniswap Support — What Is UniswapX?](https://support.uniswap.org/hc/en-us/articles/17515415311501-What-is-UniswapX). Auction mechanism, available liquidity sources and claimed mitigation benefits.
17. [CoW Protocol — Fair Combinatorial Batch Auction](https://docs.cow.fi/cow-protocol/concepts/introduction/fair-combinatorial-auction). Order matching, solver competition and protections.
18. [CoW Protocol — Solvers](https://docs.cow.fi/cow-protocol/concepts/introduction/solvers). Solver participation, settlement conditions and permissions.
19. [CoW Protocol — Coincidence of Wants](https://docs.cow.fi/cow-protocol/concepts/how-it-works/coincidence-of-wants). How some offsetting orders avoid external AMM liquidity.
20. [Uniswap Developers — Aggregator Quote API](https://developers.uniswap.org/docs/api-reference/aggregator_quote). Exact-input and exact-output slippage-parameter differences.
21. [Uniswap Developers — Swap API FAQs](https://developers.uniswap.org/docs/trading/swapping-api/faqs). Scope of permits, approvals and quote-specific signatures.

## Editorial Methodology, Calculation Assumptions and Disclosure

**Reporting cutoff: November 7, 2026.** This is a durable original educational guide rather than a report of a specific on-chain incident. All numerical pool examples are modeled, not live. We do not recommend any particular route, exchange or tolerance percentage and make no claim to have measured actual trades on Terenval Wallet.

**Core AMM assumptions:** An idealized, full-range, fee-adjusted constant-product pair starts at exactly **1,000,000 USDC and 400 ETH**, or **100,000 USDC and 40 ETH** in the shallow comparison. The initial marginal benchmark is **2,500 USDC per ETH** and the simulated fee is **0.30% of input**. USDC is assumed exactly $1 and ETH is valued at $2,500 for illustrative dollar reference comparisons. The calculation uses `output = ETH_reserve × (USDC_input × 0.997)/(USDC_reserve + USDC_input × 0.997)`; amounts are computed at full floating-point precision and displayed with stated rounding. No fees, gas, MEV, taxes, bridge or token-behavior effects are hidden inside the price-impact formula except the explicitly modeled 0.30% pool swap fee. We avoid claiming that every Uniswap v3 or v4 pool is covered by this simplified model.

**Slippage and execution assumptions:** The minimum-output table applies a percentage floor to the **already-priced, fee-adjusted quote** of a conventional exact-input swap. It expresses worst-permitted additional deterioration, **not** a predicted fill, guarantee of success, or all-in loss cap. Dollar equivalents assume ETH remains at the example reference price of $2,500. Actual apps may use different UI conventions and exact-output calculations. MEV or sequential-swap examples are model illustrations, not evidence that a real user was sandwiched.

**Original analytical contribution:** Terenval's Quote-to-Receipt Ledger, Depth Stress Grid, Slippage Exposure Budget, MEV Exposure Gate and Route Decision Framework are original explanatory frameworks intended to improve user comprehension and auditability. They are not conventional technical metrics established by Uniswap, not pre-deployed Terenval software capabilities, and not guarantees of protection, settlement or favorable pricing.

**Native Terenval boundary:** Terenval News and Terenval Answers are part of the ecosystem that also develops Terenval Wallet. This educational discussion refers to the practical benefits and limits of self-custody, token authorization and network support, but **does not claim** Terenval Wallet has verified built-in DEX routing, UniswapX/CoW support, dynamic slippage settings, MEV detection or private transaction protection. We do not claim any endorsement or affiliation from Uniswap, Ethereum.org, CoW Protocol or other cited organizations. Content is educational only and is **not financial, legal, security or tax advice**; crypto trades and contract interactions can cause partial or total loss of economic value.

## Editorial & technical review

- Published by: Terenval
- Technical review: Terenval Wallet team
- Last reviewed: 2026-11-07

## Sources

- [Uniswap Support — Price Impact vs Price Slippage](https://support.uniswap.org/hc/en-us/articles/8643794102669-Price-Impact-vs-Price-Slippage)
- [Uniswap Support — What Is Price Impact?](https://support.uniswap.org/hc/en-us/articles/40074715860365-What-is-price-impact)
- [Uniswap Developers — Understanding Swaps, Execution and Safety Parameters](https://developers.uniswap.org/docs/get-started/concepts/traders/swaps)
- [Uniswap Support — How to Change Slippage on the Web App](https://support.uniswap.org/hc/en-us/articles/8643879653261-How-to-change-slippage-on-the-Uniswap-Web-app)
- [Uniswap Web App — The Swap Screen Explained](https://support.uniswap.org/hc/en-us/articles/39862756339341-Uniswap-Web-App-The-Swap-Screen)
- [Uniswap Developers — Uniswap v2 Pricing, External Quotes and Risk Checks](https://developers.uniswap.org/docs/protocols/v2/concepts/pricing)
- [Uniswap Developers — v2 SDK Slippage, Minimum Output and Deadline](https://developers.uniswap.org/docs/sdks/v2/guides/swapping)
- [Uniswap Developers — v3 Single Swaps, Minimum Output and Price Limit](https://developers.uniswap.org/docs/protocols/v3/guides/swapping/single-hop-swapping)
- [Uniswap Developers — v3 Multi-hop Routing and Minimum Output](https://developers.uniswap.org/docs/protocols/v3/guides/swapping/multi-hop-swapping)
- [Uniswap Developers — Fee Structures Across v2, v3 and v4](https://developers.uniswap.org/docs/get-started/concepts/fees)
- [Uniswap Developers — Uniswap v4 Hooks and Custom Pool Logic](https://developers.uniswap.org/docs/get-started/concepts/hooks)
- [Uniswap Developers — Dynamic Fees in Uniswap v4](https://developers.uniswap.org/docs/protocols/v4/concepts/dynamic-fees)
- [Uniswap Developers — Protocol Glossary: Price Impact and Slippage](https://developers.uniswap.org/docs/get-started/concepts/glossary)
- [Ethereum.org — Maximal Extractable Value and Sandwich Trading](https://ethereum.org/developers/docs/mev/)
- [Uniswap Support — UniswapX Uses Different Slippage Settings](https://support.uniswap.org/hc/en-us/articles/20071671202189-Can-I-change-slippage-when-using-UniswapX)
- [Uniswap Support — What Is UniswapX?](https://support.uniswap.org/hc/en-us/articles/17515415311501-What-is-UniswapX)
- [CoW Protocol — Fair Combinatorial Batch Auctions](https://docs.cow.fi/cow-protocol/concepts/introduction/fair-combinatorial-auction)
- [CoW Protocol — Solvers and Order Settlement](https://docs.cow.fi/cow-protocol/concepts/introduction/solvers)
- [CoW Protocol — Coincidence of Wants](https://docs.cow.fi/cow-protocol/concepts/how-it-works/coincidence-of-wants)
- [Uniswap Developers — API Quote and Slippage Parameters](https://developers.uniswap.org/docs/api-reference/aggregator_quote)
- [Uniswap Developers — Permit2 and Transaction Approval Scope](https://developers.uniswap.org/docs/trading/swapping-api/faqs)

Official article: https://wallet.terenval.com/answers/what-is-slippage-in-crypto/

[Machine-oriented Terenval index](https://wallet.terenval.com/llms.txt)
