Crypto Markets Have a Hidden Risk Core

One of the foundational principles of portfolio management is diversification. By spreading their investments across different assets, investors can reduce the impact of any single investment on their portfolio’s overall performance.

The idea here is pretty simple: assets that do not move in the same direction at the same time can help cancel out each other’s losses during periods of market stress.

In crypto, diversification has been used as a solution to volatility, but this practice often translates into a habit of buying more coins. While a portfolio with twenty tokens may feel safer than one with two coins, the effectiveness of diversification in crypto is a subject of much debate, and for good reason.

Digital assets tend to become highly correlated during periods of market turbulence. A new academic study, “Core–periphery analysis of risk dependence among cryptocurrencies¹,” shares some insight into the topic, suggesting that the safety intuition from diversification may largely be an illusion.

According to research, understanding the relationships among cryptocurrencies may be more important than simply increasing the number of tokens held in a portfolio.

By mapping how tail risk actually moves across more than 200 cryptocurrencies over four years, the study shows that the market is organized around a small, shifting “core” of assets that transmit shocks to everyone else, while a much larger periphery absorbs risk without generating it.

Holding more tokens does not change this structure; it just means holding more correlation and, during stress, more of the same risk.

To understand why this matters, let’s take a deeper look into what diversification is actually designed to accomplish and how the critical assumption that ensures its success becomes particularly important when evaluating diversification in crypto markets.

Why Diversification Exists and Where It Breaks Down

In the world of investment, diversification is a way to manage risk. It is a key risk management strategy that spreads investments across a mix of assets, industries, or markets.

The goal here is to limit exposure to any single asset so that a single negative event doesn’t destroy your entire portfolio. This is what professionals call diversifying away idiosyncratic risk, the risk specific to a single company or asset, as opposed to risk that affects the entire market.

Studies have shown that a well-diversified portfolio, with about 25 stocks, provides the most cost-effective risk reduction.

By combining assets with different performance drivers, diversification helps reduce a portfolio’s volatility, improve risk-adjusted returns, and enhance resilience during market downturns.

So, if one of your holdings falls while another rises or even holds steady, losses in the first investment get cushioned by the second, smoothing your path of total wealth over time. The practice of diversification also lowers the odds that an investor will be forced to sell at the worst possible moment.

Diversification is not without limitations, though. For starters, spreading your capital too thin weakens your upside. Not only can overemphasizing diversification dilute your returns, but it also increases complexity, adds costs through transaction fees and brokerage commissions, and even creates a false sense of security if the assets in your portfolio remain highly correlated.

The thing is, the benefits of diversification can only hold if the assets in a portfolio respond differently, preferably in opposing ways, to market influences. Diversification simply can’t eliminate systematic risk, the kind that hits an entire asset class or the broader economy at once, no matter how many names are in the portfolio. That’s because during systemic crises, even assets that usually move independently may begin moving together, reducing the benefits of spreading out your investment funds.

A great example of systemic risk is COVID-19, which had a far-reaching global impact. Almost no asset class, industry, or economy was spared its severe effects. What diversification can reduce is unsystematic risk, which is specific to an investment, such as risks related to a company based on the nature of its business and its financial health.

Overall, diversification is only as good as the correlation assumptions behind it; if assets that looked independent in calm markets suddenly start moving together under stress, the protection you thought you had can vanish exactly when it is needed most.

In traditional markets, there are several ways to diversify a portfolio. One way is to diversify across asset classes, with each carrying its own risks and opportunities. Different asset classes include stocks, commodities, bonds, real estate, cash, short-term cash equivalents, and exchange-traded funds (ETFs).

When investing in public equities, you can further diversify into growth stocks, which are stocks of a company whose earnings and revenue rise faster than its industry average, and value stocks, which are stocks of a company trading at a lower price than its current fundamentals would suggest.

Another way is to invest across sectors such as healthcare, semiconductor manufacturing, finance, agriculture, oil, green energy, aerospace, and automobiles. Investors may also make decisions based on the market capitalization of an asset or company. The choice here is between large-cap and small-cap stocks, with each offering a different size, approach to raising capital, and growth potential.

Once an investor has decided which diversification strategy to pursue, they must determine what percentage of the portfolio to allocate to each choice.

The Diversification Illusion in Crypto

Now, what does diversification, or “not putting all your eggs in one basket,” look like in crypto?

Well, crypto is highly volatile, which is both its major risk and its biggest opportunity. To mitigate this volatility, meaning the price of a digital asset can experience large swings in either direction within minutes, investors diversify their crypto portfolios.

One way they achieve this is by adding new tokens. These additions can be chosen based on market cap. Bitcoin and Ether are two major assets, with BTC being the only one with a trillion-dollar market capitalization.

Given the tens of thousands of coins available, investors can consider different types, such as stablecoins, governance tokens, utility tokens, and others. One can also diversify across sectors such as DeFi, memes, gaming, file storage, privacy, AI, and DePIN, as well as consensus mechanisms like PoW and PoS.

That’s not all. Crypto investors have other options too. There are tokenized assets, which are digital representations of real-world assets such as art, bonds, or real estate, and crypto stocks such as Coinbase (COIN ), MARA Holdings (MARA ), and Strategy (MSTR ).

IDOs, IEOs, and presales, meanwhile, offer a way to add “small-cap” assets to a crypto portfolio. Investors have ETFs too, which offer the benefits of easy, direct access through traditional brokerage accounts, eliminating the need to manage digital wallets, private keys, or crypto exchanges.

While there are many options, the challenge of diversification in cryptocurrency markets is that many investors equate it with holding more tokens. In practice, owning more cryptocurrencies doesn’t necessarily create a more diversified portfolio.

That’s because if all the assets in a portfolio are exposed to similar market forces, i.e., investor sentiment, liquidity conditions, and systemic shocks, there’s a high possibility they will all behave similarly when markets become stressed.

We saw this happen in October 2025, when the entire crypto market experienced a drawdown. At the time, BTC (BTC )and ETH (ETH ) lost 11% and 13%, respectively, while midcap and small-cap digital assets declined far more, in some cases by more than 50%. The infamous 10/10 event, triggered by US President Donald Trump’s 100% tariffs on Chinese imports, marked the crypto market’s “most devastating liquidation event in history, wiping out over $19 billion in leveraged positions within 24 hours.”

But was this an outlier? Not really. We have seen this happen time and again. When the Bitcoin price declines, altcoins get obliterated.

The new study provides evidence for this, showing that connections among cryptocurrencies tend to strengthen during periods of market stress, reducing diversification opportunities. As contagion spreads across the market, the benefits of simply holding more tokens can diminish substantially.

Interestingly, this isn’t limited to crypto alone. Historically, Bitcoin has shown low correlation to traditional assets like stocks and bonds, making it an attractive addition to the diversified portfolios of institutional allocators.

But the relationship between Bitcoin and equities has grown noticeably tighter over the past few years, particularly during periods of macroeconomic stress, eroding some of that diversification benefit at the portfolio level.

Inside the crypto market itself, the bigger problem is correlation among cryptocurrencies. With risk not evenly distributed, effective crypto diversification requires understanding how cryptocurrencies are connected to one another and how risk is transmitted across the market. For investors, this means looking beyond the number of assets they hold and instead examining the structure of relationships between those assets.

Mapping How Risk Moves Through Crypto

As cryptocurrencies become an increasingly important part of global financial markets, attracting the attention of regulators, investors, and researchers alike, a growing number of studies have analyzed digital assets, which remain highly volatile and subject to sudden shifts in market conditions.

Cryptocurrency returns, as the latest study notes, are characterized by volatility clustering, heavy tails, and strong co-movements, especially during periods of market stress. In such situations, shocks can spread rapidly from one cryptocurrency to another, increasing market instability and making it difficult for traditional risk models to capture the broader dynamics of systemic risk.

As a result, more studies are now examining the crypto market as a system of interconnected assets rather than analyzing individual digital assets.

This is what the latest paper by authors Susanna Levantesi and Giulia Rotundo from the Department of Statistical Sciences, Sapienza University of Rome, and Gabriella Piscopo from the Department of Economic and Statistics, University of Naples Federico II, sets out to do, analyzing the dynamics of the core-periphery structure of tail risk dependence among 221 crypto assets between Jan. 1, 2022 and Nov. 21, 2025.

Rather than focusing only on price movements or market cap, the researchers analyze how systemic risk spreads through the cryptocurrency ecosystem over time.

From a methodological perspective, the paper proposes an integrated framework combining breakpoint detection, dynamic network construction, CoVaR estimation, and spectral core-periphery analysis to study the transmission of systemic tail risk in crypto markets.

Conditional Value at Risk (CoVaR) is one of the most widely used measures to assess the risk of the system conditional on a specific asset being under distress, and is therefore particularly suitable for capturing tail dependence in crypto.

Unlike most existing studies, which focus on static dependence structures, pairwise spillovers, or traditional centrality measures, this study tracks how the core-periphery structure changes over time and across different market regimes, providing a better understanding of how systemic importance evolves in the digital assets space.

There is currently limited evidence on how the core-periphery (CP) structure of cryptocurrency markets evolves across different market regimes, according to the latest study, meaning we still don’t know much about which cryptos consistently play a systemic role, how the composition of the core changes over time, and how these changes then influence the propagation of tail risk.

The paper first identifies structural breaks in the market to distinguish different regimes, then estimates Value at Risk (VaR), CoVaR, and other measures used to build dynamic networks of risk spillovers among cryptocurrencies. Finally, a CP model is applied to identify cryptos that are persistently central to the transmission of systemic risk and those that remain at the periphery.

The study finds that the crypto market consistently exhibits a core-periphery structure. A relatively small group of digital assets forms a highly interconnected “core” that plays a key role in transmitting systemic risk. Meanwhile, a much larger group remains on the “periphery,” exerting more limited influence on market-wide contagion.

Importantly, the size and composition of this core are not static; they evolve across different market regimes and tend to follow recurring phases of expansion, peak concentration, and reorganization.

The research also shows that periods of market stress are associated with denser networks, stronger contagion effects, and a significantly larger systemic core. During these periods, crypto assets become even more interconnected, increasing the likelihood that shocks originating in one asset will spread throughout the broader market.

Rethinking Portfolio Construction in Crypto

Using breakpoint-detection methods, the study identifies three major turning points that divide the 2022-2025 period into distinct market regimes, each with its own statistical signature.

The first regime began with the collapse of the FTX exchange on November 10, 2022. Its co-founder and CEO, Sam Bankman-Fried (SBF), is currently serving a 25-year federal prison sentence for the multibillion-dollar fraud that led to the collapse.

The shock from this event led to negative skewness and extremely high kurtosis in returns, indicating severe tail risk.

The second regime starts on January 2, 2024, a rebound phase tied to growing institutional interest, regulatory clarity, and anticipation of a spot Bitcoin ETF. This one is marked by positive average returns but still elevated tail risk.

The third breakpoint, beginning January 9, 2025, signals the onset of a high-activity period, marked by increased speculative trading and heightened sensitivity to global macro-financial conditions, reflecting a shift toward more aggressive risk-taking and stronger cross-market spillovers. Early 2025 showed more symmetric returns and lower kurtosis, but late 2025 again exhibited the negative skewness and fat tails associated with market turmoil.

The study says:

“Overall, these breakpoints can be interpreted as endogenous responses of the cryptocurrency market to major financial disruptions, regulatory and macroeconomic turning points, and innovation-driven cycles. They highlight the strong interaction between external events and the internal evolution of risk dependence structures.”

This reinforces the idea that the crypto market’s risk structure isn’t driven by random noise but by identifiable shocks.

So what happens inside the resulting risk networks? In every regime, the study finds that the market organizes into a small core of highly interconnected cryptocurrencies that act as the primary transmitters of systemic tail risk, surrounded by a much larger periphery that absorbs risk but rarely generates it.

This structure, however, is not static. During the FTX-driven crisis, the core expanded from roughly 15 cryptos to 39 at its peak, as contagion pulled more assets into tight co-movement. The 2023 recovery period produced the broadest core and the densest network observed in the entire sample, thanks to renewed optimism and anticipation of ETFs. By contrast, the 2024 regime was more fragmented and unstable, and by 2025, the network had become more regular and consolidated, a pattern interpreted as a sign of a maturing market structure.

Most importantly, only a limited number of cryptos have been persistently central over time; most assets move between the core and the periphery depending on market conditions.

This means that systemic importance “changes across different market regimes.” A token may not be important in one period but can become very influential in another. According to the study:

“To conclude, we highlight that assets that are in the periphery in one phase can become relevant in another. This means that the risk management of cryptocurrency portfolios must be dynamic and not only based on a fixed list of tokens.”

For investors, this means you can’t just create a diverse crypto portfolio and be done with it.

Moreover, assets that seem different on the surface may still be tightly linked. Some peripheral assets, meanwhile, may offer greater diversification benefits.

Genuine diversification in crypto means identifying assets that respond to different drivers and recognizing that within any given market regime, large parts of the altcoin universe may simply be leveraged bets on the same underlying trade.

Meanwhile, tracking the evolving core can provide regulators and exchanges with an early-warning tool to monitor which cryptocurrencies are becoming systemically important enough to warrant closer oversight.

Conclusion

An important concept in financial planning and investment management, diversification helps lower portfolio risk by spreading wealth across different assets. And it is as relevant in cryptocurrency markets as it is in conventional finance.

But buying more coins as a diversification strategy doesn’t meaningfully reduce risk, because diversification isn’t about the quantity of assets but about their behavior when markets come under stress. Current market data on crypto correlation shows that large parts of the altcoin universe move together with Bitcoin, and that co-movement tends to intensify exactly when diversification matters most.

This means portfolio construction, risk management, and regulatory attention must remain dynamic rather than static to keep pace with the rapid evolution of crypto risk structures.

References

1. Levantesi, S., Piscopo, G. & Rotundo, G. Core–periphery analysis of risk dependence among cryptocurrencies. Physica A: Statistical Mechanics and its Applications, 131750 (2026). https://doi.org/10.1016/j.physa.2026.131750