This is where the bridge either holds or it doesn’t. Six lessons built the whole multi-chain machine: how lost atomicity turns an arb into a carry trade, how bridges move value across ledgers, the exposure window and the inventory, latency and reorg risks it reopens, the inventory and rebalancing that lock up your capital, the trust models and tail risk that can erase years of spread in one exploit, and the shared sequencers and cross-domain MEV racing to buy atomicity back. No pretests now — just the exam, asking whether it stuck. The trap is usually the option that’s 90% right.
How this exam works
This is a graded exam. Questions arrive one at a time. Once you submit an answer it is final — there is no going back, no retry, and a wrong answer simply fails that question. Your score stays hidden until the very end, where you need 70% to pass. Read every option twice before you commit.
A trader says cross-chain arbitrage is 'the same as on-chain arbitrage, just on two chains.' What is the sharpest correction?
Select an answer to continue.
Course Recap
Big picture
Cross-chain arbitrage, end to end
- Cross-chain arbitrage end to end
- From atomic to at-risk
- Atomicity gave no-inventory-risk and flash loans
- No transaction spans two ledgers
- The non-atomic CEX-DEX cousin
- A carry trade in an arbitrage costume
- How bridges move value
- A bridge is a cross-chain messenger
- Lock-and-mint backs a wrapped token
- Burn-and-mint moves the canonical asset
- Liquidity networks pay from inventory
- Always two transactions on two ledgers
- The risks atomicity hid
- Exposure window equals finality plus relay
- Inventory risk is directional
- Drift scales with the square root of time
- Reorg can leave a one-legged position
- Gap must clear fee plus drift plus tail
- Inventory and rebalancing
- Pre-position capital on both chains
- Same-direction flow drains one side
- Capacity is capped by inventory
- Rebalancing costs a fee and time
- Judge it on return on capital
- Bridge trust and tail risk
- The bridge is now your counterparty
- Trusted, optimistic, light-client, liquidity
- Ronin, Wormhole, Nomad honeypots
- A thin carry against a fat tail
- Cap per bridge and diversify
- Shared sequencers and cross-domain MEV
- MEV relocates, it does not vanish
- Centralized sequencers hold ordering power
- Shared sequencers enable atomic cross-rollup
- Bonded relayers, escrow, intents approximate atomicity
- From atomic to at-risk
Key Takeaways
What to remember
- Lost atomicity is the whole story. A single-chain arb is one all-or-nothing transaction; cross two chains and it splits into two legs on two ledgers, reintroducing inventory, latency, settlement and tail risk. Cross-chain arb is the non-atomic, inventory-heavy cousin of CEX–DEX arbitrage — a carry trade wearing an arbitrage costume.
- Every bridge is two transactions and a trust assumption. Lock-and-mint backs a wrapped token with a locked original (drain the lock and it depegs); burn-and-mint moves the canonical asset; liquidity-network bridges pay you from pre-funded inventory. All settle as source tx → finality → relay → destination tx, never atomically.
- The exposure window is where the money leaks. It equals the finality wait plus the bridge relay; during it you hold a directional position, and expected drift grows with the √ of the window. Deeper confirmations cut reorg risk but lengthen the window — a 1% gap nets ~0.73% after a ~0.05% fee and ~0.22% drift, while a 0.3% gap never clears.
- Capital is the real constraint. With no cross-chain flash loan, you pre-position inventory on both sides; same-direction flow drains one side to zero (capacity is inventory, not opportunity), and rebalancing costs a fee plus idle transit time. Judge the strategy on return on capital — it can lose to simply lending the money.
- The tail can erase years of spread. The bridge is your counterparty, and Ronin (~625M), Wormhole (~326M) and Nomad (~190M) show how concentrated value behind novel code gets drained. A 2% annual catastrophe on 4M is ~80,000 expected per year, bimodal and ruinous — so cap per-bridge exposure, diversify, and size with fractional-Kelly humility.
- MEV relocates, and atomicity is the prize. Cross-domain MEV is ordering power across chains; centralized sequencers hold it locally, shared sequencers can restore atomic cross-rollup execution at the cost of centralization, and bonded relayers, escrow swaps and intents each approximate the atomicity you lost — paying with trust, latency, capital, or centralization.