Correlation Matrix
The Correlation Matrix shows the statistical relationship between all 31 supported pairs on the Hanzo AI platform — a number between -1.0 and +1.0 that tells you how closely two pairs move together. A correlation of +1.0 means the pairs move in perfect lockstep. A correlation of -1.0 means the pairs move in perfect opposition. A correlation near 0 means they move independently. Use the Correlation Matrix to avoid stacking correlated risk, to identify hedge pairs, and to understand which instruments are currently behaving as risk-on or risk-off proxies.
What this page covers
- How to read correlation scores between trading pairs
- How to avoid correlated risk by identifying pairs that move together
- How to use negative correlation for hedging
- How the timeframe selector affects correlation readings
What Is Pair Correlation?
Correlation measures how consistently two instruments move in the same direction over a given period. A high positive correlation (above +0.80) means that when one instrument goes up, the other almost always goes up too — and by a similar magnitude. A high negative correlation (below -0.80) means they reliably move in opposite directions.
For traders, correlation has two critical implications. First, it reveals hidden risk: if you have long positions on EURUSD, GBPUSD, and AUDUSD simultaneously, and all three are highly correlated, you don't have three trades — you have one large USD bet with three times the exposure. Second, it reveals hedging opportunities: a negative correlation pair can reduce drawdown by partially offsetting losing positions.
Reading the Correlation Matrix
| Correlation range | Classification | Trading implication |
|---|---|---|
| +0.90 to +1.00 | Perfect positive correlation | These pairs are essentially the same trade. Never hold both in the same direction — you are doubling risk with no diversification benefit. |
| +0.70 to +0.89 | Strong positive correlation | The selected sample shows substantial movement overlap; document the lookback and reassess as conditions change. |
| +0.40 to +0.69 | Moderate correlation | Partial overlap but meaningful independence. Can trade both, but be aware that a shared macro event (e.g., NFP) will move both simultaneously. |
| -0.10 to +0.39 | Low or no correlation | Largely independent. These pairs provide genuine diversification — a loss on one is not likely to be replicated on the other. |
| -0.40 to -0.69 | Moderate negative correlation | Partial hedge. Holding both in opposite directions partially offsets directional risk. |
| -0.70 to -1.00 | Strong negative correlation | Reliable hedge relationship. These pairs are the basis for systematic hedging strategies — when one rises, the other reliably falls. |
Key Correlations to Know
Some correlation relationships may recur in particular samples or regimes. The examples below are historical reference ranges, not stable relationships, hedges, or position-construction instructions:
- EURUSD and GBPUSD: a historically positive relationship in some samples — compare the selected lookback and note regime changes.
- XAUUSD and USDJPY: a relationship that can be positive or negative across samples — inspect the selected period rather than assuming a hedge.
- AUDUSD and NZDUSD: often compared as risk-sensitive currencies; the displayed coefficient varies by period.
- EURUSD and USDCHF: frequently compared as dollar-related series; the coefficient is not fixed and does not prohibit any holding.
- XAUUSD and DXY (US Dollar Index): often compared as gold and dollar series; the relationship can change during stress events.
Timeframe Selector
Correlations are not static — they shift with market regimes, economic conditions, and risk sentiment. The Correlation Matrix's timeframe selector lets you view correlations over different lookback periods:
| Lookback period | What it shows | Best used for |
|---|---|---|
| 1 week | Current intraday/intraweek correlation — often noisy | Scalpers and day traders checking same-session correlation |
| 1 month | Medium-term trend correlation — reflects current regime | Swing traders building a multi-position book |
| 3 months | Seasonal and macro correlation — regime-level relationships | Position traders and long-term bias construction |
| 1 year | Longer-sample correlation — still subject to regime changes | Reviewing historical pair relationships |
Pre-Session Correlation Risk Checklist
Use this correlation review as a paper-study exercise to identify overlapping historical behavior. It does not prescribe exposure, hedging, or position limits:
- List every pair in today's watchlist
Write out the full list: e.g., 'EURUSD long, GBPUSD long, XAUUSD long, USDJPY short'. Even pairs you 'might' trade should be included.
- Check all pairs against each other in the matrix
For each pair combination in your list, check the 1-month correlation. Flag any pair combination showing > +0.75 correlation. For your example above: EURUSD/GBPUSD is likely > +0.85 — flag it.
- Compare flagged pairs
For pairs with a high displayed correlation, record the coefficient and lookback period. Treat the result as overlapping historical movement, not a prescribed sizing rule.
- Check the shared factor
If several instruments share a currency or macro factor, document the possible overlap and the limits of pair-to-pair correlation. Do not infer aggregate exposure from the matrix alone.
- Record study assumptions
Document the selected pairs, lookback period, data source, and hypothetical assumptions. No account budget, concurrent-position limit, or risk percentage is implied.
Using Correlation With Alert Intel
Alert Intel can display notifications when readings from related pairs appear together. Treat correlated-pair co-occurrence as a study case, not an entry trigger or reliability ranking.
| Alert setup | Logic | Why it works |
|---|---|---|
| EURUSD BOS detected + GBPUSD Volume reading | BOS on one pair appears with a volume reading on the related pair | Compare the two readings and later candles |
| XAUUSD Wyckoff Spring + USDJPY bearish BOS | A gold phase label appears with a USDJPY structure label | Record whether the historical context agreed or diverged |
| AUDUSD breakout + NZDUSD Volume reading | A breakout label appears with a volume reading on the related pair | Compare shared-factor context without inferring direction |
Three Practical Use Cases
- Review a historical sample
Open the Correlation Matrix and list pairs for a selected lookback. Record coefficients, regime notes, and whether later behavior matched the historical relationship.
- Compare a hypothetical hedge
For paper study, compare a pair with a negatively correlated series and document the assumptions and limitations. Do not treat the example as a hedge instruction.
- Compare related breakout labels
If one pair has a breakout label, inspect a related pair's displayed behavior and record agreement or divergence without assigning a probability.
Frequently asked questions
How frequently does the Correlation Matrix update?
The matrix updates every 4 hours — at each 4H candle close. Intraday correlation shifts are typically noise; the 4H update cadence balances recency with stability.
Can I view the correlation between any two specific pairs?
Yes. Click any cell in the matrix to open a detailed two-pair correlation view showing the correlation coefficient over all four timeframes, a scatter plot of returns, and historical correlation trend over the past year. This is useful for understanding whether a pair relationship is strengthening or weakening.
Is correlation available for all supported pairs?
Yes — the matrix covers all 31 supported pairs that Hanzo AI monitors. The full 31×31 matrix would be unwieldy to display, so the default view shows a curated selection of stronger positive and negative relationships. Use the search bar to find a specific pair quickly.