Financial Analyst/ Guest author, RADEX MARKETS
Lee W. is a seasoned professional trader with over 10 years of experience. Passionate about sharing valuable expertise and unique market insights, Lee W. now serves as an external and independent market analyst for RADEX MARKETS.
SHARE
Two currency pairs that normally move almost in step suddenly begin drifting apart. One races ahead, the other lags behind, and a trader spots an opportunity to profit if they eventually come back into line.
Pairs trading is a market-neutral strategy that involves going long on one closely correlated asset and short on another. The aim is to profit from the gap between them, known as the spread, returning to its usual range, rather than trying to predict whether the wider market will rise or fall.
I've always found that appealing because it shifts the focus away from making heroic market predictions and towards measuring a relationship. This guide explains how pairs trading works, why traders use it, how to select a suitable forex pair, and how experienced traders fine-tune and manage the strategy. Grab a coffee and let's begin.
Pairs trading is a market-neutral strategy involving two historically correlated assets. A trader buys the underperformer and sells the outperformer, expecting the spread between them to return to its normal range.
The potential profit comes from their relative performance, not from correctly predicting the market's overall direction.
Ready to explore a strategy built around relative value rather than outright market direction?
Pairs trading is a market-neutral, relative-value strategy in which a trader simultaneously buys one asset and sells another closely correlated asset. The objective is to profit when the spread between the two returns to its historical range, rather than from predicting the direction of the wider market.
The strategy is considered a form of statistical arbitrage because the trading decision is based on a measurable relationship between two markets. It also uses mean reversion, although it is the spread between the assets that is expected to revert, not necessarily the price of either asset on its own.
Read more: What is arbitrage trading? How beginners profit from it
Read more: What Is Mean Reversion and How Can Traders Use It?
The traditional version of pairs trading often involves two shares from the same industry. In forex, however, the same principle can be applied to two highly correlated currency pairs, such as EUR/USD and GBP/USD.
These pairs frequently move in a similar direction because they both measure a European currency against the US dollar. If GBP/USD suddenly outperforms EUR/USD by an unusually large amount, a pairs trader might sell GBP/USD and buy EUR/USD, expecting the difference between their movements to narrow.
In my experience, this is where the strategy becomes particularly interesting. You are not simply deciding that the dollar will rise or fall, a prediction that markets occasionally enjoy making look rather foolish. Instead, you are asking whether the relationship between two currency pairs has moved too far away from its normal trading range.
For this reason, pairs trading is best understood as trading the relationship between two markets rather than trading either market in isolation.
Pairs trading works by tracking the spread between two markets that have historically moved together.When that spread moves unusually far from its average, the trader buys the underperforming asset and sells the outperforming one, expecting the two to converge.
A simple way to express the spread is:
Spread = Price A − Price B
Some traders use a price ratio instead:
Spread = Price A ÷ Price B
In practice, forex prices often need to be standardised or weighted before they are compared because different currency pairs have different price levels and volatility. The basic principle remains the same: measure the relationship, identify an unusual divergence and trade the expected return towards normality.
Think of the two markets as a motorway and its service road. They follow roughly the same route, but the terrain may occasionally push them further apart. Provided they are still heading towards the same destination, the two roads eventually move closer together again. The spread measures the distance between them.
Pairs trading is described as market-neutral because the long and short positions are intended to offset much of the broader market movement.
If both currency pairs rise, the long position may gain while the short position loses. If both fall, the opposite may happen. The result depends mainly on whether the gap between the pairs narrows, not on whether the entire forex market moves higher or lower.
"Market-neutral" does not mean risk-free or perfectly hedged. If the positions are sized badly or the historical relationship breaks down, the trade can still become directional and lose money.
Suppose EUR/USD and GBP/USD have been moving closely together, but GBP/USD suddenly climbs much faster. A pairs trader may:
I prefer to think of this as trading the elastic band between the two markets. A stretched band may snap back, but if it keeps stretching, there is usually a reason, and blindly adding to the position is rarely a clever response.
Ultimately, the success of the trade depends on convergence between the two legs, not on correctly calling the direction of the US dollar or the wider forex market.
Pairs trading appeals to traders because it does not depend entirely on predicting the market's next direction. In simple terms, the strategy can:
However, pairs trading is not a holy grail. It only works as intended while the relationship between the two markets remains reasonably stable, and even a well-researched pair requires disciplined risk management.
Choosing the pair is usually the most important, and most time-consuming, part of the process. Get this stage right and everything that follows becomes easier; choose two markets simply because their charts look vaguely similar and the strategy can unravel rather quickly.
High correlation means that two assets have historically moved in the same direction to a similar degree. Traders commonly look for a strong, stable positive correlation of approximately 0.80 or higher, although the appropriate threshold depends on the timeframe and method being used.
EUR/USD and GBP/USD are a familiar forex example. Both pairs often respond similarly to changes in US dollar sentiment, so their price movements can be closely correlated.
Correlation should be measured across multiple periods rather than accepted from a single snapshot. A reading of 0.90 today is useful, but it does not guarantee the relationship will remain that strong next week.
A shared driver is an economic or market influence that helps explain why two assets move together. Correlation is generally more convincing when there is a genuine reason behind it rather than a relationship that has appeared by chance.
EUR/USD and GBP/USD share the US dollar as their quote currency. Both may therefore respond in a similar way to Federal Reserve policy, US inflation figures and changes in demand for the dollar.
Other pairs can be connected through related economic forces. AUD/CAD, for example, is influenced by commodity sentiment, although the currencies do not respond identically: Australia has strong links to metals and Chinese demand, while Canada is particularly sensitive to oil.
A shared driver does not guarantee that the relationship will continue, but it gives the correlation a more credible foundation.
Cointegration tests whether the spread between two markets has remained relatively stable and mean-reverting over time. This differs from correlation, which only measures how closely their prices have moved together.
Two currency pairs can show a high correlation while their spread steadily drifts further apart. Cointegration is designed to identify a longer-term equilibrium relationship, making it especially useful when testing whether a divergence has a realistic chance of reversing.
Research by Carrasco Blázquez, De la Orden De la Cruz and Prado Román concluded that cointegration was the most efficient of the methods they examined for structuring a pairs trading strategy. The study compared several approaches, including correlation, distance and cointegration, although no statistical method can guarantee a profitable trade. (Pairs Trading Techniques: An Empirical Contrast, 2018)
For beginners, the key distinction is simple: correlation asks whether two markets move together, while cointegration asks whether the spread between them tends to return to a stable relationship.
Liquidity refers to how easily an instrument can be traded without causing a meaningful change in its price. Because pairs trading involves opening and closing two positions, both currency pairs need sufficient liquidity and consistently competitive spreads.
Poor liquidity can create:
This is one reason major currency pairs are often more practical candidates. EUR/USD and GBP/USD generally offer deeper liquidity than exotic pairs, although spreads and execution conditions can still change around news releases and quieter trading hours.
A pair is only genuinely tradeable when the relationship is statistically credible, and both legs can be entered and exited efficiently.
Read more: What is liquidity in Forex and how is it measured
Method |
What it measures |
Advantages |
Limitations |
|---|---|---|---|
| Correlation | How closely two prices move together | Simple and intuitive; useful as an initial filter, often at approximately 0.80 or above | Does not prove the spread is stable or mean-reverting; correlation can break |
| Distance | The gap between two standardised price series | Objective and relatively easy to use when identifying closely moving markets | Can be less effective when the spread itself is trending |
| Cointegration | Whether the spread has a stable, mean-reverting relationship | Focuses directly on the spread; widely used in more advanced pairs analysis | More complex and normally requires a statistical test such as the augmented Dickey–Fuller test |
The following example is hypothetical and provided for illustration only. It does not represent live market prices, guaranteed results or a recommendation to trade these currency pairs.
Suppose EUR/USD and GBP/USD have maintained a rolling correlation of approximately 0.90. Over the period being analysed, the ratio between their prices has normally remained close to 1.175:
Spread ratio = GBP/USD price ÷ EUR/USD price
GBP/USD then begins outperforming EUR/USD, pushing the ratio well above its recent average.
Stage |
EUR/USD |
GBP/USD |
Spread ratio |
|---|---|---|---|
| Historical relationship | - | - | Approximately 1.175 |
| Trade entry | 1.0800 | 1.2960 | 1.200 |
| Trade exit | 1.0900 | 1.2840 | Approximately 1.178 |
At the entry point, the ratio has widened from its usual level of around 1.175 to 1.200. The trader interprets this as an unusually large divergence and opens two positions:
The pairs then begin converging. EUR/USD rises to 1.0900 while GBP/USD falls to 1.2840, bringing the ratio back to approximately 1.178, close to its historical range.
The trader closes both positions:
This is a particularly tidy illustration because both legs make money. Real pairs trades are rarely so obliging. One position may lose while the other gains more, and the trade can still produce a positive net result if the spread moves in the expected direction.
I find it more useful to judge these trades by the combined position than to become emotionally attached to either leg. Closing the profitable side early while leaving the losing side open would dismantle the hedge and turn a relative-value trade into an unplanned directional bet.
The potential profit in this example comes from the two currency pairs returning to their usual relationship, not from correctly predicting the direction of the wider forex market.
A pairs trading strategy becomes far more useful when it is turned into a repeatable process. As you become more proficient with this strategy the easier it will become. The following six steps take the idea from pair selection through to the final exit while reducing the temptation to improvise halfway through the trade.
Setting your selection criteria means deciding what must be true before two currency pairs qualify for consideration. This is often the hardest and most time-consuming stage because it provides the framework for every decision that follows.
Before searching for a trade, define:
Establishing these rules in advance helps prevent traders from adjusting the criteria merely because a particular chart looks tempting.
A candidate list is a shortlist of currency-pair combinations that meet your initial requirements. You can screen the market manually or use trading software, a spreadsheet or a statistical model to compare multiple relationships.
For example, a trader might begin with liquid major and minor pairs before filtering for correlations above 0.80. Any pair that passes the first screen can then be tested more thoroughly for spread stability and cointegration.
How frequently the list needs updating depends on the strategy. A trader holding positions for several weeks may screen less often, while someone using shorter timeframes will need to regenerate candidates more regularly.
Applying additional analysis means examining the shortlisted pairs before committing capital. The statistical relationship identifies a possible opportunity, but technical and fundamental analysis can help refine the entry, profit target and stop level.
A trader might look at:
This extra layer should confirm the setup rather than provide an excuse to ignore the original rules.
Executing a pairs trade means opening the long and short positions as close together as possible. Both legs should ideally reach the intended prices or target spread without leaving the trader exposed to an unexpected directional move.
If one position fills while the other is delayed, the trader temporarily holds what is known as legging risk. Instead of having a hedged relative-value position, they have a naked long or short trade until the second order executes.
Forex traders do not usually face the share-borrowing constraints associated with short-selling stocks. The main concerns are timing, liquidity and fill price. Limit orders and, where supported by the trading platform, linked or OCO-style order management may help coordinate execution, although an OCO order does not automatically guarantee that both legs will fill together.
Slippage on each position may appear small, but pairs trading starts and ends with two legs. Every additional cost reduces the available edge.
Managing a pairs trade means monitoring the combined position against the rules established before entry. The spread is the main focus, but traders must also watch for news or economic developments that could affect either leg.
If the spread reaches its target earlier than expected, the trader may:
Psychology becomes important here. It is tempting to give a losing spread "a little more room" or hold a winning trade indefinitely because it seems to be behaving nicely. Markets have a habit of becoming less polite once the original plan has been abandoned.
Closing the trade means exiting both positions when the spread reaches the profit target, hits the stop-loss level or exceeds the maximum holding period. With clear rules, the mechanics are usually straightforward; following those rules is the more difficult part.
One of the classic mistakes is refusing to close a modest loss after the trade has exceeded its expected duration. The trader assumes that waiting another few days will allow the relationship to recover, but a slow or absent reversion may indicate that market conditions have fundamentally changed.
I have found that a time-based exit can be just as valuable as a price-based stop. If the reason for taking the trade has not played out within a sensible period, continuing to hold it is often hope dressed up as patience.
A repeatable pairs trading process therefore depends on objective selection, coordinated execution and the discipline to close both legs when the original conditions no longer apply.
Once you have the basic process in place, a clean pairs trade comes down to three measurable judgements. This is where "the charts look as though they have drifted apart" becomes "the spread has moved far enough for me to act."
Each measurement addresses a different weakness in the basic strategy.
A z-score measures how many standard deviations the current spread sits above or below its historical average.
The basic formula is:
Z-score = (Current spread − Mean spread) ÷ Spread standard deviation
A trader first chooses a lookback window, such as the previous 20 to 60 bars. The rolling average and standard deviation of the spread are then calculated across that period.
A straightforward rules-based approach might be:
Suppose the spread between EUR/USD and GBP/USD reaches a z-score of +2. This indicates that it is two standard deviations above its recent mean. The trader sells the relative outperformer and buys the underperformer, then closes both positions if the z-score returns towards zero.
However, a z-score moving beyond +3 or −3 should not automatically be treated as a better bargain. It may indicate that the underlying relationship is breaking down due to a policy shift, economic surprise or broader change in market conditions. Adding repeatedly to a widening spread can turn a controlled trade into a rather expensive lesson.
A hedge ratio determines the relative position size of each leg. Market neutrality does not necessarily mean placing the same monetary amount on both sides; it means balancing the positions so their exposure to shared market movements is as close as reasonably possible.
The ratio may be estimated using:
Suppose GBP/USD has been approximately 1.3 times as volatile as EUR/USD over the chosen period. Using a simplified volatility adjustment, the GBP/USD position might be reduced to roughly 77% of the EUR/USD exposure:
1 ÷ 1.3 = 0.77
If the strategy calls for buying EUR/USD and selling GBP/USD, the trader will therefore use a smaller position on the more volatile GBP/USD leg. The exact lot sizes would also need to account for pip value, account currency and the trader's maximum permitted loss.
The hedge ratio can change as volatility and market relationships evolve. It should be recalculated periodically because an outdated ratio may allow what began as a balanced trade to acquire an unintended directional bias.
The half-life is an estimate of how long the spread normally takes to travel halfway back towards its mean. It provides a practical measure of mean-reversion speed and can be estimated from the spread's mean-reversion coefficient using a model related to the Ornstein–Uhlenbeck process.
The mathematics can become fairly involved, but the trading idea is straightforward. Half-life can help a trader:
Imagine that the EUR/USD and GBP/USD spread has an estimated half-life of three trading days. The trader would expect a meaningful portion of the convergence to occur within that general period.
If the spread remains wide after several half-lives, the original assumption deserves serious reconsideration. Holding the trade for three weeks because "it normally reverts in three days" is not patience; it is evidence that the model may no longer describe the market.
Used together, the z-score, hedge ratio and half-life turn pairs trading into a more structured decision process by defining the entry, balancing the exposure and placing a sensible limit on the time allowed for convergence.
Experience with pairs trading usually changes what you pay attention to. Beginners tend to concentrate on spotting a wide spread; after doing it for a while, you spend far more time asking why the spread widened and whether the relationship is still valid. This is a common mistake traders make.
These are the risks I would keep firmly on the watchlist:
Read more: What Is the Spread in Forex? Learn to Calculate and Trade It
Pairs trading rewards careful research, regular monitoring, clear exit rules and the discipline to act on them. It may be market-neutral, but it is certainly not set-and-forget.
Ready to explore correlated forex pairs while managing both sides of the trade from one account?
A pair in pairs trading consists of two related assets traded simultaneously. The trader buys the relative underperformer and sells the outperformer, aiming to profit if their historical price relationship returns to normal.
A pairs trading strategy is a non-directional, relative-value approach that identifies two historically related assets whose spread has moved outside its usual range. The trader goes long on the underperformer and short on the outperformer, expecting the spread to converge.
If EUR/USD and GBP/USD normally move together but GBP/USD begins significantly outperforming, a trader might buy EUR/USD and sell GBP/USD. Both positions are closed if the spread returns towards its historical average.
Yes. The main risk is that the historical relationship permanently changes and the spread continues widening instead of reverting, potentially creating a loss across the combined position.
Yes, pairs trading can be applied to highly correlated currency pairs such as EUR/USD and GBP/USD or AUD/USD and NZD/USD. The principle is the same as with stocks: trade the expected reversion of the spread rather than the direction of one market.
Pairs trading is on the more advanced side because it requires an understanding of correlation, short-selling, position sizing and the costs of managing two legs. Beginners should learn the fundamentals of risk management first and practise the process on a demo account, with no assumption that profits are guaranteed.
There is no universal minimum, but two simultaneous positions both require margin and incur trading costs. Position sizes should reflect your risk tolerance, and the account should retain a sufficient margin buffer for adverse price movements.
Pairs trading offers a market-neutral way to seek profit from the changing relationship between two correlated assets rather than from predicting the market's overall direction. Its potential edge comes from careful pair selection, coordinated execution and measurable controls such as z-scores, hedge ratios and half-life estimates.
Research and discipline matter far more than making a dramatic market call. Start by testing the relationship, define the conditions that would prove the idea wrong and practise managing both legs as a single position.
Lee W. is a seasoned professional trader with over 10 years of experience. Passionate about sharing valuable expertise and unique market insights, Lee W. now serves as an external and independent market analyst for RADEX MARKETS.
Reviewed by RADEX MARKETS
Risk Warning: Trading derivatives and leveraged products carries a high level of risk, including the risk of losing substantially more than your initial investment.
Trading strategies:
Related concepts:
Trading derivatives and leveraged products is high risk and may result in losses exceeding your initial investment. Read our T&C and PDS before trading.
OPEN ACCOUNT