Currency correlation trading means using the statistical relationship between two forex pairs to decide which trades to take, which to skip and how much total risk you are really carrying. Most traders meet correlation the hard way: two “different” trades lose together, and a 1% risk plan turns into a 2% loss. Used properly, correlation is less a signal generator and more a risk filter that keeps your open positions honest.
Below we cover how correlation is measured, how to read a matrix, what the dollar index does and does not tell you, and a simple routine you can test on a demo account.
TL;DR
- Correlation runs from -1 to +1. Near +1 two pairs move together, near -1 they move opposite, near 0 they are unrelated.
- Two long positions on positively correlated pairs are often one trade wearing two names. Size them as one.
- Correlations drift. Check the matrix on more than one timeframe and never treat a number as permanent.
- The dollar index is mostly the euro in disguise, so it is a weak “second opinion” for EUR/USD.
- Treat correlation as a risk filter first. Your entry still needs price action and confluence.
Table of Contents
- What currency correlation actually measures
- Typical relationships between major pairs
- How to read a correlation matrix
- Why direction changes what the number means
- Using correlation as a risk filter
- Hedging with correlation: useful, but not free
- The dollar index and forex correlation
- Gold, commodities and other relationships
- A simple forex correlation strategy to test
- Common mistakes in currency correlation trading
- Perspective: what 18 years of live trading taught us
- Learn this with a mentor and a live market
- FAQ
- Sources
What currency correlation actually measures
In forex, correlation is the statistical relationship between two currency pairs, expressed as a coefficient between -1.0 and +1.0, according to Investopedia’s guide. At +1 the pairs move in the same direction all of the time. At -1 they move in opposite directions all of the time. At 0 the relationship is random.
Two details trip up beginners:
- The sign is direction, the size is strength. A reading of -0.85 is a strong relationship, just an inverse one. It does not mean the pairs move by the same number of pips, as a UK-facing broker guide explains.
- Some platforms show percentages. The same scale is often written as -100 to +100, so +85 and +0.85 say the same thing.
The reason correlation exists at all is simple. Many pairs share a currency. EUR/USD and GBP/USD both have the dollar on one side, so when the dollar moves broadly, both tend to react. Pairs that sit on opposite sides of the dollar, such as EUR/USD and USD/CHF, tend to mirror each other.
Typical relationships between major pairs
A published one-day sample from IG’s trading guide, dated 26 November 2019, gives a feel for the pattern:
| Pair A | Pair B | Reading in that sample | Relationship |
|---|---|---|---|
| EUR/USD | GBP/USD | +0.81 | Strong positive |
| GBP/USD | USD/JPY | +0.83 | Strong positive |
| USD/JPY | EUR/JPY | +0.86 | Strong positive |
| EUR/USD | USD/CHF | -0.54 | Moderate negative |
| EUR/USD | USD/CAD | -0.72 | Strong negative |
Treat these as an illustration of how a matrix looks, not as numbers to trade from. They are a single day from several years ago, and that is exactly the point of the next sections: correlations move.
The broad logic is stable even when the decimals are not. Pairs quoting the dollar as the second currency (EUR/USD, GBP/USD, AUD/USD) tend to rise and fall together. Pairs with the dollar first (USD/CHF, USD/CAD) tend to run against them.
What about GBP/JPY and USD/JPY?
Readers often ask whether GBP/JPY and USD/JPY move together. They share the yen, so yen strength or weakness pushes both in the same direction, and they commonly show a positive relationship. The strength of it depends on whether the move is driven by the yen or by sterling. If the Bank of England is the story, GBP/JPY can separate from USD/JPY quickly. Check the live number on your own platform rather than assuming.
How to read a correlation matrix
A correlation matrix is a grid with the same pairs down the side and across the top. Each cell holds the coefficient for that pair of pairs. A broker’s correlation matrix guide sets out the scale: +100 means two symbols moved identically, -100 means exactly opposite, 0 means independently.
Three reading rules save time:
- It is symmetrical. The cell for EUR/USD against GBP/USD equals the cell for GBP/USD against EUR/USD. You only need half the grid.
- The diagonal is useless. Each pair compared with itself is always +1.
- The window matters. A matrix is calculated over a chosen timeframe and number of bars, for example H1 over the last 100 bars. Change the window and you change the answer.
That last point is where most mistakes happen. A 100-bar H1 reading describes roughly the last four to five trading days. A daily reading over six months describes something much slower. Investopedia notes that a six-month trailing correlation gives a clearer perspective because correlations shift with sentiment and global economic factors.
Our habit is to look at two windows: a short one for what is happening this week and a long one for the underlying relationship. When they disagree, we trust neither blindly and reduce size.
Why direction changes what the number means
A correlation figure only becomes useful when you add the direction of your trades. The same guide spells it out:
- At +90: two longs, or two shorts, tend to make or lose money together. A long and a short tend to offset.
- At -90: a long and a short tend to make or lose together. Two longs, or two shorts, tend to offset.
So going long EUR/USD and short USD/CHF is, in practice, close to doubling one bet on the euro against the dollar. Going long EUR/USD and long GBP/USD is also close to doubling one bet on a weaker dollar. Going long EUR/USD and short GBP/USD is partly a hedge, and partly a bet on euro versus sterling.
None of this makes those trades wrong. It means you must know which one you are actually placing.
Using correlation as a risk filter
The most practical use of correlation is not finding trades. It is limiting the number of trades that are secretly the same. The matrix approach is to identify strong correlations among current or planned positions and keep correlation between open positions low, because positions can either duplicate the same price action or cancel each other out.
A worked example
Say you risk 1% per trade on a £10,000 account, so £100 per trade. You spot a long setup on EUR/USD and another on GBP/USD, and the two pairs are correlated at roughly +0.80.
- If you treat them as separate, you are risking £200, or 2%, on one idea: the dollar falling.
- If both stop out together, you lose 2% from what felt like two independent trades.
The fix is a hard rule, not a feeling. Either take the better of the two setups at full size, or take both at half size so combined risk stays at 1%. Our guides on risk management and position sizing in practice and the 2% risk rule show how to set that cap in cash terms.
A simple correlation checklist before you click buy
- List every open position and every pending order.
- Note the base and quote currency of each, and your direction.
- Check the matrix on two windows for any pair that shares a currency.
- If two trades have a reading above roughly +0.7 or below -0.7 and lean the same way on the shared currency, count them as one position.
- Size the combined exposure to your single-trade risk limit.
The 0.7 threshold is a working rule of thumb we use, not a standard. Pick your own number, write it down and stick to it.
Hedging with correlation: useful, but not free
Correlations can be used to add exposure in the same expected direction or to hedge risk, but IG’s analysts warn that a wrong forecast can magnify losses and that an apparently protective hedge can turn out to be ineffective. A hedge built on a +0.8 relationship still has a 20% gap in it, and that gap widens exactly when markets get stressed.
There is also a behavioral side. A hedge can make a losing position feel safe, so traders hold on longer than their plan allows. Winning trades have their own trap: after a run of profits, traders start treating gains as spare cash and take extra correlated risk, a pattern described as the house money effect. Both habits quietly inflate risk beyond what the matrix shows.
If hedging is part of your toolkit, read our guide to hedging in forex first. It covers when a hedge reduces risk and when it only adds cost and complexity.
The dollar index and forex correlation
The US Dollar Index (DXY) measures the dollar against six currencies. TradingView lists the weights as euro 57.6%, Japanese yen 13.6%, pound sterling 11.9%, Canadian dollar 9.1%, Swedish krona 4.2% and Swiss franc 3.6%.
That weighting explains most of what traders see:
- EUR/USD and DXY are strongly negative. With the euro at well over half the index, a falling EUR/USD usually means a rising DXY. This is close to arithmetic rather than a market insight.
- USD/JPY and DXY lean positive. A stronger dollar against the yen lifts the index, though the yen’s smaller weight means the link is looser.
- GBP/USD and DXY lean negative, with sterling at under 12% of the basket.
The trap is using DXY as confirmation. As one analysis puts it, DXY to confirm a EUR/USD trade can mislead because the index largely reflects the same currency movement. A genuine second opinion has to come from something not already inside the first.
DXY is more useful for pairs with a small weight in the basket, like AUD/USD or NZD/USD, or as a quick read of broad dollar mood before you decide which dollar pairs deserve attention. For a structured way to combine such reads, see our piece on daily market bias, where three signals must align before we trade.
Gold, commodities and other relationships
People often ask which pair correlates with gold. Gold is priced in dollars, so XAU/USD frequently moves inversely to the dollar and can show a positive link with EUR/USD. Like everything above, that relationship comes and goes. If you trade gold alongside dollar pairs, treat them as related exposure and size accordingly. Our XAUUSD trading strategy guide walks through the risk math.
Commodity currencies such as CAD and AUD often respond to oil and metals, but the relationship is looser than the textbook says. Verify before you build a rule on it.
A simple forex correlation strategy to test
Correlation is a poor stand-alone entry signal. Because relationships change, a strategy built purely on “pair A lags pair B” tends to fall apart when conditions shift. A sturdier approach uses correlation in a supporting role.
Step by step
- Find the setup first. Build your idea from price action, structure and higher-timeframe context. Our multi timeframe analysis guide shows how.
- Check correlation second. Does the trade overlap with something already open? If yes, cut size or pass.
- Look for confluence across independent factors. Another correlated pair agreeing is not independent evidence. A key level, a session timing and a structural break are. See confluence trading.
- Pick the cleanest expression. If the dollar view shows up in three pairs, choose the one with the clearest structure and the tightest stop.
- Log it. Record the correlation reading, window and outcome in your journal so you can see over 30 or more trades whether the filter actually helps.
A demo test works well here. Our 30-trade EUR/USD demo plan is a good template: run it once with no correlation filter and once with, then compare drawdowns, not just profits.
Timing matters too
Correlation tends to tighten when a currency is driving the market, such as around a major dollar data release, and loosen in quiet periods. London and New York overlap hours often produce the cleanest dollar-driven moves, which is why the trading session overlap guide is worth reading alongside this one.
Common mistakes in currency correlation trading
- Treating a number as a law. A coefficient is a snapshot. Recheck it.
- Using one short window. A hot week can make unrelated pairs look linked.
- Calling a double position diversification. Long EUR/USD plus long GBP/USD is rarely diversified.
- Hedging out of fear. If you are hedging to avoid taking a stop, that is tilt, not strategy.
- Ignoring execution. Opening several correlated trades at once makes slippage and spread costs stack. Our notes on trade execution best practices help here.
- Forgetting trading costs and leverage. More open positions mean more margin in use, and correlated losses arrive together.
Perspective: what 18 years of live trading taught us
We have watched the same pattern repeat across more than 18 years in live markets. A trader has a losing week, and when we review the journal, the problem is rarely the entries. It is three or four trades that were the same idea in different clothes. The account did not lose on three bad setups. It lost on one, three times.
That is why we teach correlation as part of risk management and decision quality, not as a standalone system. Fancy matrices do not fix a plan that has no hard rule for total exposure. A clear rule does: one idea, one risk budget.
We also see how losing streaks distort judgment. After a few stops, the brain stops trading the chart and starts trading the pain. Correlation checks give you a calm, mechanical step to run before you click, which is exactly when you need one.
If you are new, start by simply counting how many currencies you are actually exposed to. If you are experienced, test whether your correlation rules improve your drawdowns over 30 trades. Either way, keep it simple.
— Gabriel
Learn this with a mentor and a live market
Reading about correlation is the easy part. Applying it while a trade is moving against you is harder. At Trader Gibkey we teach price action, risk management and trading psychology in a structured path, with live sessions and a community where you get feedback on real trades.
Options on our site include the Price Action (Core) plan at £149 per month, New Generation Liquidity at €349 per year, and Diamond (Lifetime Access) at €3499 one-off. If you would rather have someone review your own trades and exposure, mentorship starts at €149 for a single session, with 3 sessions at €399 and 5 sessions at €559. For daily market reads, there is a Telegram channel at €49 per week or €109 per month.
Trading forex carries a real risk of loss, and no course or correlation filter removes it. Start on a demo account and risk only money you can afford to lose.
FAQ
What is currency correlation trading?
It is using the statistical relationship between currency pairs to manage and shape trades. The coefficient runs from -1 to +1. Most traders use it to avoid duplicate exposure rather than to generate entries.
Is the dollar index a good confirmation for EUR/USD?
Not really. The euro makes up 57.6% of the index, so DXY mostly repeats the EUR/USD move. Look for independent confirmation such as structure, key levels or timing.
How often should I check a correlation matrix?
Before every new trade that shares a currency with an open position. Look at a short window and a longer one, since correlations change over time.
Sources
- Investopedia: using currency correlations to your advantage
- Broker guide to currency correlations and forex correlation pairs
- Broker correlation matrix guide (PDF)
- Guide to trading currency pair correlations in forex
- TradingView: U.S. Dollar Index (DXY) chart and index info
- US Dollar Index (DXY): what the basket actually measures
- House money effect for traders