CEX-to-DEX Arbitrage
Learn how centralised and decentralised exchange prices are arbitraged across order books and AMMs while accounting for gas, MEV, bridge, inventory and failed-transaction risk.
Reading progress — saved on this device
CEX-to-DEX arbitrage compares a centralised order book with an on-chain AMM or DEX route and trades when the executable price gap exceeds all costs and settlement risks.
Learning objectives
- Compare CEX VWAP with DEX executable output rather than displayed spot prices.
- Include gas, DEX fees, MEV and transaction failure in net edge.
- Use inventory and hedge design to avoid waiting for cross-rail transfers.
What it is
A DEX price is generated by pool reserves, liquidity curves and routing. A CEX price comes from resting orders. Differences arise because the markets update asynchronously and capital is segmented.
Professional arbitrage tends to hold inventory on both sides or hedge rapidly rather than move each asset across venues for every trade.
How it works
DEX quote size matters: AMM price impact can make a visible 1% gap vanish for a large order. Aggregators may split routes across pools, introducing more contracts and gas.
Transactions enter a mempool or private orderflow depending on chain and route. Other actors can compete, back-run or sandwich poorly protected swaps; price can change before inclusion.
A DEX transaction can revert, leaving the CEX hedge filled but the on-chain leg absent. The system therefore needs a contingency for leg risk.
Cross-chain opportunities are more complex because bridges create latency and smart-contract/custody risk. A token with the same ticker on two chains may not be economically identical if representations differ.
How to analyse and apply it
| Check | Why it matters | What to verify |
|---|---|---|
| DEX quote at size | Captures AMM price impact. | Use exact token amount, route and slippage tolerance. |
| CEX depth | Provides hedge price. | Calculate same-size VWAP. |
| Inclusion risk | Can create leg mismatch. | Model gas, priority, private routing and revert scenarios. |
| Inventory/bridge | Determines rebalance risk. | Prefer pre-positioned inventory where feasible; map bridge dependencies. |
A strategy is not complete until the signal, sizing, execution, invalidation and review process are explicit. Any discretionary override should be recorded so it can be separated from the tested rule set.
Worked example and thought exercise
A token can be bought on a CEX for £100.00 and sold on a DEX quote at £101.20 for a £50,000 size, suggesting 1.2%. At that size DEX price impact reduces proceeds by 0.45%, combined fees are 0.25%, and gas/expected MEV another 0.15%.
Net theoretical edge is only about 0.35% before leg failure and inventory costs. If the transaction reverts after the CEX buy, the trader is left long the token.
Thought exercise: why is a DEX screen price particularly misleading for a large arbitrage order?
Common mistakes and practical workflow
- Comparing CEX last price with DEX spot display.
- Ignoring gas and MEV because the headline spread is large.
- Assuming both legs execute atomically when they do not.
- Bridging every trade instead of analysing pre-funded inventory.
Practical workflow
- Request size-specific DEX route and CEX VWAP.
- Subtract all deterministic costs and a buffer for adverse selection.
- Plan leg order and failure hedge.
- Use slippage and transaction controls appropriate to the chain.
- Rebalance inventory only through verified settlement paths.
✅ Knowledge checkpoint
- Why must DEX price be measured at the intended size?
- What is leg risk in a CEX-to-DEX trade?
- How can a reverted transaction create directional exposure?
- Why do bridges make cross-chain arbitrage qualitatively riskier?
FAQs
❓ Is on-chain arbitrage atomic?
Some multi-DEX on-chain strategies can be atomic, but a CEX leg cannot generally be part of the same blockchain transaction.
❓ What is MEV risk here?
Other participants may reorder or trade around visible transactions, worsening execution or taking the opportunity.
❓ Why pre-fund inventory?
It reduces dependence on slow transfers during the opportunity window.
❓ Are same-ticker tokens on different chains identical?
Not always; wrappers, bridges and redemption claims can differ.
📋 Summary
CEX-to-DEX arbitrage is an execution-engineering problem as much as a pricing one. The trade only exists after size-specific price impact, fees, gas, MEV, leg failure and settlement dependencies are included.
Want this in a personalised order?
Take the crypto assessment and get a custom path of 10 modules matched to what you already know. Free, no card required.
Build my path →