How to Calculate Total Risk for Multiple Crypto Positions? Correlation and Portfolio Heat

Trading Strategies

01 October 2026

Putting ten coins into the same account does not mean you have ten independent risks. If they rise and fall together in the same market cycle, what you actually hold may just be one exposure split into ten parts. A measurement in September 2026 showed that the average pairwise correlation of twelve mainstream crypto assets over 90 days was 0.61, and over the most recent 30 days it rose to 0.64. Correlation is rising, and the protection from diversification is getting weaker.

OKX Exchange
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！

Total risk is not the sum of individual position risks

Many people calculate total risk by evaluating each coin's risk separately and then adding them up. This method is already not accurate enough for stock portfolios, and the deviation is even larger for crypto assets.

The reason is the linkage between positions. Academic research provides a specific comparison: in an analysis of fourteen crypto assets, risk factor concentration reached 67.89%, meaning nearly 70% of portfolio risk was driven by a few common factors. By contrast, a diversified stock portfolio usually has this number between 30% and 40%. How many coins you hold and how many things determine your risk are two different questions.

BTC and ETH are the most typical example. In the 90-day measurement above, Ethereum's correlation with the overall market reached 0.89, and over the most recent 30 days it rose further to 0.93. Cardano rose from 0.77 to 0.90, and XRP rose from 0.87 to 0.90. The meaning of these numbers is direct: when Bitcoin moves sharply, these assets follow by nearly 90%. Holding them at the same time does not make the portfolio more stable in extreme conditions.

When is correlation most likely to fool you

Correlation is dynamic, and its direction of change works against holders.

Gate's glossary materials point out that crypto asset correlation tends to rise during periods of violent market volatility, causing seemingly diversified multi-asset portfolios to fall together under stress. In a bull market "everything rises," and in a bear market "everything falls." The effect of diversification is weaker in both directions than in calm periods.

The September 2026 measurement confirms this: the 30-day correlation (0.64) is higher than the 90-day correlation (0.61). This means that in just the past month, synchronicity within the crypto market has increased. If you set your position ratios in early summer and have not reviewed them since, you are relying on outdated assumptions.

There is one exception. Monero's correlation in the measurement above was only 0.33, making it the only asset in the group whose daily volatility was mainly explained by its own factors. But you need to see where this low correlation comes from: Monero faces regulatory pressure, and multiple trading platforms in Europe have delisted it. These events drive its price in ways unrelated to the overall market. Low correlation here does not mean "safe asset"; it means "idiosyncratic risk." An asset with low correlation may have its own risk that is larger than the market risk it helps you diversify away.

How to calculate your actual exposure

The following steps do not require statistical software. Doing them in order will give you a judgment that is more reliable than intuition.

Step 1: Write down the weight of each position. Not the quantity, but the percentage of total portfolio market value. This step often already reveals the problem: if Bitcoin and Ethereum together account for 70%, and the remaining eight coins split 30%, then the risk of your portfolio is mainly determined by the first two positions. The correlation level of the other eight coins has little impact on the whole.

Step 2: Find shared connections beyond price. Risks not captured by correlation numbers are often more deadly than correlation itself. Check: Which assets run on the same chain? Which assets are custodied by the same platform? Which assets depend on the same stablecoin? An analysis article on CoinMarketCap gives a concrete scenario: a single failure at a custody platform will hit all positions stored on that platform at the same time, no matter how out of sync their prices usually look.

Step 3: Do a manual stress test. Take the largest single-day percentage drop of the overall market in the past twelve months, and apply that same percentage to all your positions to calculate the total portfolio value. This method will overestimate losses on low-correlation positions, but precisely because of that "overestimation," it gives you a floor: if you cannot accept even this floor, the portfolio size is too large.

Step 4: If you are using leverage, recalculate the distance. When holding multiple positions in the same margin account, high correlation means all positions will enter loss territory at the same time during market volatility. Collateral is not consumed proportionally; it is eroded by all positions simultaneously, and the liquidation price approaches faster than the result from single-position calculations. What you need to look at is not how far each position is from liquidation, but how far the entire account is.

Portfolio heat: a more direct signal than correlation

Correlation tells you how positions move together, but it does not directly tell you "whether it is too crowded right now."

A plainer way to judge is to look at the funding rate. Gate's risk analysis materials point out that the funding rate of perpetual contracts is a holding cost or income that needs to be included in portfolio return and risk; when the average funding rate falls, the risk of a pure long strategy becomes higher. The funding rate reflects how much leveraged longs are willing to pay to maintain their positions. The higher the rate, the more crowded the longs are. Once the price reverses, there is more fuel for a stampede of liquidations.

Another signal is exchange net inflow. Funds flowing into trading platforms usually mean preparation to sell, while outflows mean holders are moving assets out of exchanges. Large inflows do not necessarily cause a decline, but they are a leading indicator of potential selling pressure, and it is worth checking before you adjust portfolio weights.

A simple portfolio heat check can be combined like this: Is the 30-day correlation among your main assets higher than the 90-day correlation? If yes, and the funding rate is at a recent high, then the "crowdedness" of your current portfolio may be rising. This does not mean you should reduce positions immediately, but it means that continuing to hold at current weights carries more risk than when you set those weights.

One thing easy to overlook in practice

Do not rely only on the "portfolio risk" number provided by a trading platform. Different platforms have very different margin models and risk parameters. The same set of positions may be safe on platform A but close to the liquidation line on platform B. If you have positions on multiple platforms, you need to check the margin status on each platform separately. You cannot use the health of one platform to infer another.

If your assets are spread across multiple on-chain wallets and exchange accounts, manual aggregation becomes harder. The display of tools like DeFiGuard illustrates one direction: after connecting wallet addresses, it automatically pulls multi-chain holdings and calculates a correlation matrix and risk contribution charts. Such tools are currently mainly in English and oriented toward overseas platforms, so the barrier for Chinese users is not low. Without a suitable tool, maintaining a position weight table manually in a spreadsheet and updating the rough direction of correlation once a month (checking whether major assets have risen and fallen together over the past 30 days) is better than relying entirely on feeling.

Completion standard

Figuring out total risk is not about getting a precise number, but about being able to answer three questions:

First, in extreme market conditions, roughly how much could my portfolio fall at worst? Answer with the floor number from the stress test, not with "it is fine because I diversified."

Second, which positions contribute the most to my risk? Look at weights, and also look at shared connections. An asset with only 5% weight but placed on the same chain as your main position may have an actual risk contribution far higher than 5%.

Third, if correlation rises from the current level to 0.90, can my position size still withstand it? The crypto market tends toward high correlation during stress periods. Recalculating with a 0.90 assumption is closer to the scenario you need to prepare for than calculating with 0.60.

OKX Exchange
A leading global cryptocurrency platform,suitable for both beginners and experienced traders.
New user benefit: 20% off trading fees upon registration!！