Bitcoin price discovery · Part 3 of 3
A price difference between exchanges can look like free money. In practice, stale data, execution speed, fees, liquidity and settlement risk determine whether an apparent arbitrage opportunity can actually be captured.
What latency means in Bitcoin trading
Latency is elapsed time between an event and a later response or observation. There is no single “Bitcoin latency.” Several different delays matter, and confusing them can lead to incorrect conclusions.
Market-data latency
Time between an exchange event and the update reaching a trader or price provider.
Order latency
Time for an order to reach an exchange, pass checks and enter the matching engine.
Execution latency
Time until the exchange matches the order and reports the result.
Settlement latency
Time and uncertainty involved in moving assets or receiving required blockchain confirmations.
How stale prices create false opportunities
Assume exchange A appears to quote Bitcoin at $70,000 and exchange B at $70,200. Before calling the $200 difference an opportunity, a trader would need to know whether both quotes are current, whether they are bids or asks, and how much quantity is available at those prices.
A delayed ticker might show B’s old last trade after its actual bid has already fallen. A REST request can also return snapshots taken at slightly different times. Even a live stream can suffer network delay, message loss or local processing backlogs. Reliable systems use exchange timestamps and sequence information where available rather than comparing only the arrival time on one computer.
Gross spread versus executable result
A realistic comparison starts with prices that can be executed: the ask on the venue where Bitcoin would be bought and the bid on the venue where it would be sold. The gross difference must then cover every relevant cost and risk.
The calculation is path-dependent. If the buy executes but the sell does not, the trader is left with market exposure. If a transfer is required before selling, the price can change while the transaction awaits confirmations or an exchange credits the deposit.
Why professional arbitrageurs pre-position funds
Moving Bitcoin after observing a price gap is often too slow. A professional cross-exchange strategy may keep both cash and bitcoin on multiple venues so the buy and sell orders can be submitted nearly together. Balances are rebalanced later.
This reduces immediate blockchain settlement delay, but creates other exposures: capital is tied up, funds are entrusted to exchanges, withdrawals can be paused, and balances can become uneven. Faster execution does not remove counterparty or operational risk.
How speed affects liquidity and price discovery
When new information moves a leading venue, fast traders may update or cancel quotes elsewhere before slower orders arrive. This helps prices incorporate information quickly, but it also creates adverse-selection risk for market makers whose resting quotes become outdated.
Research on fragmented markets shows that latency design can change where informed orders go, how liquidity is distributed and which venue contributes to price discovery. More speed is therefore not simply “better” for every participant; it changes competition between liquidity providers and takers.
Can an individual trader use latency?
An individual can use an understanding of latency defensively: avoid treating a delayed widget as an executable quote, inspect bid and ask prices, use limit orders where appropriate, check depth and fees, and be cautious during fast markets. Those practices can reduce misunderstandings, but they do not guarantee better execution.
Trying to compete directly in latency arbitrage is different. Professional participants can use colocated infrastructure, direct market-data feeds, automated risk controls and capital distributed across venues. A public API and ordinary internet connection generally do not provide the same information or execution speed.
Practical risks that a backtest can miss
- Using last-trade prices instead of executable bid and ask prices.
- Assuming unlimited size at the best price.
- Ignoring partial fills and the risk that only one side executes.
- Applying today’s fee schedule to historical periods.
- Ignoring withdrawal suspensions, minimums and confirmation policies.
- Comparing timestamps generated by unsynchronized systems.
- Excluding failed orders, downtime and API rate limits.
- Ignoring counterparty, custody, tax and regulatory constraints.
Sources and further reading
- Building Trust Takes Time: Limits to Arbitrage for Blockchain-Based Assets
- Bank of Canada: Order-Flow Segmentation, Liquidity, and Price Discovery
- Fragmentation, Price Formation, and Cross-Impact in Bitcoin Markets
This is a technical explanation, not a recommendation to trade or an assurance that arbitrage is profitable. Trading and keeping funds on exchanges can result in loss.