Why prices differ between exchanges

Look at the same coin on two different exchanges at the same moment and you will often see two slightly different prices. This is normal, and it is not a glitch. Each exchange runs its own separate market, and a price is only ever the result of buyers and sellers meeting on that specific venue. Understanding why the numbers diverge explains a lot about how fragmented modern markets really are.

Each exchange has its own order book

The most fundamental reason is that there is no single, universal price for an asset. Every exchange maintains its own order book, filled with its own users’ buy and sell orders and matched by its own matching engine. The “price” you see is just the last trade or best quote on that book. Two venues with different participants, different order flow, and different liquidity will naturally settle at slightly different levels at any given instant.

Because each book is separate, supply and demand are effectively local, and local imbalances push prices apart. If one venue happens to have more aggressive buyers at a given moment, its price ticks up relative to a venue where sellers dominate. These imbalances appear and fade constantly as orders arrive. The deeper and busier a venue, the more stable its price tends to be; a thinner venue, with less market depth, can swing further on the same order flow, widening the gap versus larger markets.

Arbitrage keeps prices close — but not identical

If prices drift apart, participants can, in principle, buy on the cheaper venue and sell on the more expensive one. This activity, called arbitrage, tends to pull the two prices back toward each other: buying lifts the cheap venue, selling presses down the expensive one. In deep, liquid, well-connected markets this happens quickly, so differences are usually small and short-lived.

But arbitrage is not free or frictionless, and the frictions are exactly why a gap can persist:

• Fees: trading fees on both venues, plus withdrawal and deposit fees, must be smaller than the price gap for closing it to be worthwhile.
• Transfer time: moving an asset between venues can take minutes and requires network confirmations, during which the price can change.
• The spread and slippage: acting on the difference means crossing the bid-ask spread and possibly incurring slippage on both sides, which eats into the gap.
• Withdrawal limits and delays: capital may be locked on one venue and slow to redeploy.

The size of the price difference that can persist is roughly bounded by the total cost of these frictions. Below that threshold, closing the gap is not worth it, so it simply remains.

Structural reasons prices differ

Beyond moment-to-moment imbalance, several durable factors keep venues apart:

• Fiat and regional access: a venue serving a particular currency or country can trade at a persistent premium or discount when moving money in and out is difficult or restricted. When capital cannot flow freely, arbitrage cannot fully equalise prices.
• Liquidity differences: large global venues and small local ones have very different depth, so their prices respond differently to the same news or flow.
• Fee and rebate structures: different maker-taker schedules subtly shape where prices sit.
• Trading pairs and denomination: a coin quoted against a fiat currency on one venue and against a stablecoin on another introduces an extra exchange rate, so the two quotes are not strictly comparable without converting.

Prices also differ simply because information and capital take time to move. Each venue updates its own book independently, and a trade or news event that shifts one market is not instantly reflected everywhere else. Participants who watch multiple venues at once react at different speeds, so for brief moments the same asset genuinely trades at different levels in different places before the prices reconverge. The better connected and more liquid the venues, the smaller and shorter these gaps; the more isolated a venue, the longer a difference can linger. None of this requires anything exotic — it is just the physics of many separate markets updating on their own clocks.

How aggregators build a reference price

Because no single venue is authoritative, price aggregators and index providers combine data from many exchanges into one reference figure. The common method is a volume-weighted average, which gives more influence to venues where more trading actually happens, so a thinly traded outlier does not distort the headline number. Good methodologies also filter out venues with unreliable or wash-traded volume and may exclude quotes that stray too far from the pack. The result is a robust summary — but, by construction, an average. Knowing how a given reference price is built tells you why it can differ from the exact number on the one book in front of you, and why two aggregators can disagree slightly even though both are “correct” by their own rules.

Why “the price” is really an average

Because there are many venues, the single figure you see on a data aggregator or price index is usually a blend — often a volume-weighted average across several exchanges. That is a sensible summary, but it means the aggregate “price” may not exactly match any individual venue where you could actually trade. When precision matters, the price that counts is the one on the specific book you are trading against, not the global average.

This is also why comparing a fill you received on one venue to a headline price from an index can be misleading: they are measuring different things. Your fill reflects one book’s liquidity at one instant; the index reflects a smoothed average across many.

A worked comparison

Consider two venues quoting the same coin, with illustrative numbers used only to show the structure. Venue A, large and busy, shows a best bid of 100.0 and best ask of 100.1 — a one-tick spread on a deep book. Venue B, smaller, shows a best bid of 99.6 and best ask of 100.0 on much thinner size. The last traded prices might read 100.05 on A and 99.8 on B, a gap of 0.25. Is that gap an opportunity? To close it you would buy on B and sell on A, but you must cross B’s wide spread, pay trading fees on both venues, move the asset between them while the price can drift, and possibly pay withdrawal costs. Once those frictions are totalled, a 0.25 gap on a 100 asset may be entirely consumed. The example shows why observable price differences are real yet often not worth acting on — and this article describes that mechanism, not a way to trade it.

Stablecoins and reference rates add another layer to this. Many crypto prices are quoted against stablecoins rather than a national currency, and a stablecoin’s own value can drift slightly from the currency it references. When one venue prices a coin in a fiat currency and another prices it in a stablecoin, part of the apparent price difference is really the stablecoin trading a touch above or below its peg. Separating the asset’s price difference from the denomination’s difference is essential to reading cross-venue quotes correctly.

What this means

Prices differ between exchanges because there is no single market — only many separate books, each with its own participants and liquidity. Arbitrage constantly works to narrow the gaps, but fees, transfer times, spreads, slippage, and access restrictions set a floor below which differences persist. The practical takeaways are mechanical, not strategic: treat each venue’s price as local, remember that a global “price” is an average that may match no single venue, and check what currency or stablecoin a quote is denominated in before comparing. This article explains market mechanics only and is not trading or arbitrage advice.