Risk Management of Market-Neutral Portfolios in Crypto:
- Tor Jørgen Vandeskog Tunsberg
- Feb 5
- 9 min read
Updated: Feb 12
A Smarter Approach for Lenders, Capital Allocators, Brokers, & Funds
What Makes a Portfolio Market Neutral?
A market-neutral portfolio aims to generate returns that are uncorrelated with overall market movements. This means that whether the market rises or falls, the portfolio's performance should remain stable (or ideally, profitable).
In traditional finance, hedge funds have long used market-neutral strategies to manage risk and enhance risk-adjusted returns. These strategies focus on capturing relative value rather than taking directional bets on the market.
However, in crypto markets, the idea of "market-neutral" has become increasingly popular. A growing number of funds and trading firms claim to operate market-neutral strategies, primarily through linear portfolios—portfolios composed of assets and derivatives where returns are a direct function of price movements without optionality effects (e.g., perpetual futures, spot-margin trades, and basis trades).
The tricky thing is; to determine how market neutral the portfolio really is. What’s more, to accurately break down the risk of such portfolios in a number that makes sense. This goes for anyone with exposure to professed market neutral trading strategies, which is increasingly important for lenders and capital allocators, brokers, exchanges using market makers more. In order to stay competitive lenders, brokers, and funds need precise risk estimates to avoid incorrect margin requirements or excessive capital buffers.
Types of Market-Neutral Strategies
Firstly let's consider a few of the common strategies that are commonly referred to as market neutral:
1. Pairs Trading (Relative Value Trading)
This involves going long on one asset while shorting another correlated asset to profit from their relative performance rather than market-wide movements.Example: If Bitcoin (BTC) is outperforming Ethereum (ETH), a trader might go long BTC and short ETH to capture their relative price shifts rather than betting on market direction.
2. Arbitrage Strategies
Arbitrage strategies involve exploiting price inefficiencies across different markets or instruments:
Exchange Arbitrage: Buying an asset on one exchange at a lower price and selling it on another at a higher price.
Funding Rate Arbitrage: Earning a premium by taking opposing positions in spot and perpetual futures markets when funding rates are mispriced.
Cash-and-Carry Trades: Going long spot and short futures when futures trade at a significant premium to the spot price.
3. Delta-Neutral Strategies (Hedged Positions in Derivatives Markets)
These strategies involve hedging an options or futures position so that the portfolio's delta (sensitivity to price movements) is near zero.Example: Holding a long BTC position and hedging it with BTC put options to protect against downside risk.
Selecting a Candidate Portfolio
This section could be a blog by itself, so I will keep it simple and give a few guidelines to portfolio selection before introducing our candidate portfolio:
Its common to anchor the strategy with major assets like BTC and ETH, given they’re relatively more stable and widely used. Further, we diversify across sectors by bringing in coins like AVAX, ADA, or YGG can expose you to various ecosystems—Layer 1 protocols, gaming tokens, or DeFi-related projects. Next we balance risk by having both long and short positions (if your strategy allows) can hedge against market swings, aiming to reduce overall volatility. Lastly we can adjust the portfolio with market conditions by shifting weightings if you see new opportunities or need to trim risk.
The key is to combine a core set of well-established coins with smaller allocations to emerging tokens you believe have upside, all while maintaining appropriate hedges or short positions if you’re going for a market-neutral or partially neutral approach. Note that because of correlations and covariance, the sum of the long positions does not have to be equal to the sum of the short positions.
A tree map is great for quickly grasping how our positions are distributed: bigger rectangles mean bigger net positions, and the colour gives away the position direction - long (green) or short (negative) on each asset. The below shows my selected portfolio.
For this exercise I will keep the portfolio static across time and rather use the historical portfolio returns to highlight how it will perform in various market conditions. Overall, the return distribution (see histogram) has a strong center near zero, with occasional peaks and dips that confirm the portfolio’s mostly market-neutral stance but still not immune to volatility in crypto markets. Further, we can see that the returns are not fully normal as we have some larger tails with returns reaching nearly 4% gains and loss on a handful of occasions. Yet yet the mean returns is 0.082% and the portfolio standard deviation is 0.009291646 which is signalling a fairly market neutral portfolio - this is crypto after all and we expect much greater movements.
Lets take a closer look at the returns and volatility. For each point in time, I estimate my portfolio’s volatility using an EWMA approach (red line). This volatility tends to drift between roughly 0.8% and 1.3% per day, though it does spike in more turbulent periods. I compared it to my actual returns (blue line), which mostly hover near zero but can jump to positive or negative territory. We see that when the predicted volatility rises, the swings in actual returns often get a bit bigger.
Assessing True Market Neutrality: The Core Challenge
While market-neutral strategies are designed to eliminate directional exposure, true neutrality is difficult to achieve in practice. Correlation shifts, liquidity constraints, execution risks, and structural inefficiencies in crypto markets introduce unintended exposure that can lead to unexpected losses.
For a linear portfolio—one composed of spot and perpetual futures—net exposure can be calculated relatively straightforwardly by assigning a sign to long and short positions. A delta-neutral portfolio may appear risk-less at first glance, with long and short positions offsetting each other in nominal terms.
Given our portfolio, this analysis focuses on statistical market-neutral strategies that rely on relative price movements, rather than absolute directional exposure. These are particularly relevant in spot and perpetual futures markets, where hedging, correlation dynamics, and execution constraints play a crucial role in determining actual risk.
Market Neutrality Metrics
I ran a series of checks to see whether my crypto portfolio is truly market neutral relative. For this I am using the Coin Market Cap (CMC) 100 index—a broad benchmark of top 100 leading crypto assets as my market factor. To start, I calculate daily percentage returns on both my portfolio and the CMC 100. Then I measure how strongly their movements line up in three ways:
Correlation – The correlation between my portfolio’s returns and the CMC 100 came in around 0.025. That’s practically zero, indicating they barely move in tandem.
Beta – This measures how much my portfolio tends to move for each 1% change in the market. The beta estimate turned out to be only 0.0088, which is not statistically different from zero. In other words, when the CMC 100 moves, my portfolio barely budges in response.
R-squared – This is a gauge of how much the index can explain my portfolio’s day-to-day returns. The result was close to 0%, confirming that the portfolio’s ups and downs come from other factors— not from general market trends.
Put it all together, and it’s clear the portfolio is almost perfectly neutral with respect to my market factor (CMC 100 index). This is exactly what I aim for in a “market neutral” strategy: profits (or losses) driven by the individual positions I choose, rather than a general rise or fall in the crypto market.
If you’re wondering why I used the CMC 100 index as my market factor, it’s because it’s one of the broader, well-regarded indexes in the crypto space. It blends many of the biggest projects into a single benchmark, making it a decent stand-in for “the crypto market.” Since the results show my returns barely correlate with it, I can be more confident that my portfolio is insulated from wide market swings—and that’s a key point for anyone looking to reduce volatility and hedge exposure in crypto.
I also put together a simple scatterplot of my portfolio’s daily returns against those of the CMC 100, and it looks like a random cloud of points—just what you’d expect when two return streams barely track each other.
However, though the long term correlation between my portfolio returns and CMC 100 index return is ~ 0, the 30-day rolling correlation sometimes swings between roughly +35% and -35%. One theory is that correlation in general increases as prices move more dramatically either direction. My portfolio therefore might pick up some market exposure when prices move dramatically— In other words when crypto hits a big rally or a swift drop, short-term correlation often spikes. It might be worth digging deeper into those high-correlation windows to see if I can adjust my positions or hedges when volatility shoots up.
For a final visual touch, I like plot the returns of the CMC 100 index on the same chart to see if these correlation bumps coincide with big market swings. It often reveals that when the market rallies or dives sharply, correlations across the board tend to jump—even for strategies that aim to stay neutral.
Whenever the CMC 100’s daily returns (bottom plot) make big moves—especially those sharper spikes—you can see the correlation temporarily drift higher or lower - though with a lag - given we are looking at 30 days rolling correlation. Overall, it confirms that the portfolio stays pretty neutral most of the time, but market volatility can cause short-lived swings in correlation.
Accurately Estimating Risks Of Market Neutral Portfolios
Ok so now what? Even though my portfolio looks pretty solid at first glance—near-zero correlation with the market and a beta that barely moves—it’s still not immune to nasty surprises. Crypto’s volatility can take off overnight, and correlations that look “neutral” can swing wildly when the market panics or overheats.
This blog set out to provide efficient risk metrics that are not prone to the so lets get to the meat.
Why Traditional VaR Often Falls Short: Most classic Value at Risk (VaR) models assume returns follow nice, neat patterns. They might work in calmer markets, but crypto is anything but calm. With sudden price jumps, fat tails, and volatility that can spike in a heartbeat, standard VaR under normal assumptions often understates the real risk. You can think you’re prepared for a “worst-case scenario” but end up seeing losses way beyond what the model predicted.
To showcase this, I have estimated two classic VaR models one assuming a normal distribution and another assuming a skewed t-distribution at 95th percentile. As you can see below, the outcome is not very good. Our model fails to capture the changes in the risk environment when volatility spikes. Thereby our losses are far exceeded and the loss threshold (the portfolio returns breaking the VaR lines) are more frequent than we expect (we should expect 5 times in a 100 days ~ 5*365 (number of days in our sample) = 18 but we see 24 times in 365 days.
A Better Way: EWMA VaR with Cornish-Fisher Expansion
EWMA helps keep an eye on recent swings by weighting newer data more heavily. In a space where yesterday’s calm can flip to today’s chaos, its a must to account for dynamic correlations.
Cornish-Fisher adjusts our risk metrics for skewed or “fat-tailed” returns. If the portfolio can get blindsided by outlier events, this helps capture that risk more accurately.
Expected Shortfall (ES) goes beyond VaR by looking at how deep the losses might get past that VaR threshold. If you’re hit with a real tail event, ES gives you a clearer picture of the damage.
Putting these pieces together—EWMA to stay agile, Cornish-Fisher to handle non-normal returns, and ES to measure extreme losses—gives a stronger toolkit for handling crypto’s wild side. After all, being “market neutral” doesn’t guarantee peace of mind when the entire market decides to move in lockstep or pull a surprise. So rather than relying on VaR models that can miss the mark, it makes sense to adopt more dynamic, tail-sensitive methods that reflect the real behaviour of crypto markets.
Lastly, we can also see how this more dynamic approach improves capital efficiency for lenders or anyone allocating capital that need to ensure reasonable buffers.
Final Thoughts
As we’ve seen throughout this discussion, market-neutral doesn’t mean risk-free. Even if your correlation and beta are near zero, the crypto environment is rife with potential hazards—short squeezes, liquidity mismatches, and sudden correlation shifts can all catch you off guard. Meanwhile, a standard VaR approach that assumes normal distributions can severely underestimate extreme events, leading to shock drawdowns when correlations break down.
Financial players like lenders, brokers, and funds need precise, reliable risk estimates. Underpricing margin requirements or holding inadequate capital buffers can be disastrous—one big market move can wipe out months of gains, or worse.
This is exactly where Crypto Risk Solutions (CRS) steps in. We’ve built a suite of risk tools tailored for the volatility, fat tails, and correlation quirks of the crypto market. Our advanced VaR model accounts for non-normal distributions, shifting correlations, and sudden volatility spikes—helping you measure tail risk more accurately.
If you’re a lender, broker, or fund manager looking to:
Strengthen margin lending decisions
Optimize capital requirements
Protect your downside from those once-in-a-blue-moon (or rather, once-every-few-months in crypto) shock events
we’d love to help. Our platform is built on the principles we’ve explored here, combining EWMA, Cornish-Fisher expansions, and expected shortfall analysis to capture the real risks hiding beneath the surface.
Interested in seeing it in action? Contact us for a demo or consultation and learn how our smarter VaR model can help you confidently navigate the ups and downs of crypto. Don’t let outdated risk assumptions leave you vulnerable—let CRS put you on the right footing in a market that never sleeps.