How Many Cryptocurrencies Should You Hold? The Diversification Math
How many cryptocurrencies should you hold? Diversification math shows why correlation caps the benefit, when another coin stops helping, and why weights matter.
Key takeaways
- With N equal coins, volatility = σ × √(1/N + (1 − 1/N) × ρ). Correlation sets a floor of σ × √ρ that no coin count can break.
- At 75% coin volatility and 0.7 correlation, five coins cut volatility to 65.4% and twenty only to 63.4%, against a floor of 62.7%.
- One stopping rule: add equal-weight coins only while each one cuts modeled volatility by at least 1 percentage point.
- Weights beat counts. Ten coins with an HHI of 1/3 carry the volatility of three equal coins, not ten.
On this page
- What does diversification actually remove?
- The formula: volatility with N coins
- Worked example: how fast the benefit runs out
- When does another coin stop helping?
- Where the count still matters: single-project failures
- Why weights matter more than the count
- The hidden cost of every extra coin
- A checklist before you add another coin
- The bottom line
- Frequently asked questions
- Sources
The honest answer is not a number. It depends on two inputs you can estimate: how closely your coins move together, and how evenly you spread money across them. Once you see the formula, you can work out where extra coins stop earning their place.
What does diversification actually remove?
Every coin’s price swings have two parts. One part is specific to the project: a failed upgrade, a hack, a team that walks away. The other part is shared by the whole market: liquidity drying up, a broad sell-off, a change in risk appetite. Diversification averages away the first part. It cannot touch the second.
FINRA puts the goal simply: diversification “reduces the risk of major losses that can result from over-emphasizing a single security.” The catch in crypto is the size of the shared part. The U.S. Financial Stability Oversight Council noted in 2022 that crypto-asset prices “have tended to be widely correlated with each other, exposing crypto-asset market participants to largely non-diversifiable risk inside the crypto-asset ecosystem.”
That is why stock-market rules of thumb don’t transfer cleanly. Investor.gov suggests owning “at least a dozen carefully selected individual stocks” to diversify a stock portfolio, but the right count depends on correlation, and that has to be worked out for the assets you actually hold.
The formula: volatility with N coins
Assume each coin has the same volatility (σ), every pair has the same correlation (ρ), and you weight them equally. Portfolio volatility is then:
Portfolio volatility = σ × √(1/N + (1 − 1/N) × ρ)
- σ is each coin’s annual volatility, for example 75%.
- ρ is the average correlation between pairs of coins, from 0 (unrelated) to 1 (identical moves).
- N is the number of coins.
As N grows, 1/N shrinks toward zero and the result approaches σ × √ρ. That is the floor: the shared risk no number of coins can remove.
Worked example: how fast the benefit runs out
Take hypothetical coins with 75% volatility each and compare two correlation levels: 0.7, where coins mostly move together, and 0.3, where they are much more independent.
| Coins held | Volatility at ρ = 0.7 | Volatility at ρ = 0.3 |
|---|---|---|
| 1 | 75.0% | 75.0% |
| 2 | 69.1% | 60.5% |
| 3 | 67.1% | 54.8% |
| 5 | 65.4% | 49.7% |
| 10 | 64.1% | 45.6% |
| 20 | 63.4% | 43.4% |
| Floor (unlimited coins) | 62.7% | 41.1% |
- Average correlation 0.7 63
- Average correlation 0.3 43
At a correlation of 0.7, the entire available benefit is only 12.3 percentage points (75.0% down to 62.7%), and five coins already capture 78.5% of it. Ten coins capture 89.1%. At 0.3 the benefit is much larger, 33.9 points, but the pattern is the same: five coins capture 74.4% of it.
When does another coin stop helping?
A practical stopping rule is to keep adding equal-weight coins only while each one cuts modeled volatility by at least 1 percentage point. The marginal effect of the Nth coin is simply the volatility with N − 1 coins minus the volatility with N.
- At ρ = 0.7: the 4th coin cuts 1.06 points and the 5th only 0.64. The rule stops at four.
- At ρ = 0.3: the 6th coin cuts 1.34 points and the 7th only 0.98. The rule stops at six.
The 1-point threshold is a choice, not a law. The point is to decide your threshold before you go looking for the next token, not after.
Where the count still matters: single-project failures
The volatility math covers everyday swings. The count earns its keep against a different event: one project collapsing on its own, through a hack, a failed launch or a team that disappears. With equal weights, the damage from one coin going to zero is simply 1/N of the portfolio.
| Equal-weight coins | Loss if one goes to zero |
|---|---|
| 4 | 25% |
| 5 | 20% |
| 10 | 10% |
| 20 | 5% |
This is the real case for holding more than a handful of coins, and it depends on weights too: a coin at 40% that fails costs 40%, however many others you own.
Why weights matter more than the count
The formula above assumes equal weights. Real portfolios are rarely equal, and there is an exact version for any weights, as long as σ and ρ are the same across coins:
Portfolio volatility = σ × √(HHI + ρ × (1 − HHI))
HHI is the sum of squared weights. With equal weights it equals 1/N, which gives back the first formula. Now suppose you hold ten coins, but one sits at 55.8% and the other nine at about 4.9% each. The HHI is 1/3, so at ρ = 0.7 volatility is 67.1%: the same as three equal coins, and three points higher than the 64.1% of ten equal coins. Our guide to concentration risk and the HHI shows how to calculate the HHI for your own holdings.
Cash is different again. A stablecoin sleeve reduces total volatility in proportion to its weight rather than through correlation, which is why stablecoin allocation is a separate decision from the number of coins you hold.
The hidden cost of every extra coin
Each new position adds work and cost that the volatility math ignores:
- Fixed fees bite small positions. A $5 withdrawal fee is 2.5% of a $200 position but 0.5% of a $1,000 one. Comparing crypto fees fairly shows how to put every cost on the same footing.
- More to secure. Extra chains mean extra wallets, token approvals and places for a mistake.
- More to monitor. Every project can change its token supply, governance or roadmap.
- More to record. Each coin multiplies the transactions you need to track for cost basis.
If the next coin cuts modeled volatility by 0.3 points and adds all of the above, it is fair to ask what it is really doing in the portfolio.
A checklist before you add another coin
- Can you say in one sentence what role it plays that your current holdings don’t?
- Has it tended to move differently from what you already own, based on your own correlation check?
- Would the position be large enough that fixed fees stay small relative to its size?
- Will it pull weight away from an oversized holding, or just add a sliver at the bottom?
- Can you store it securely with tools you already use?
The last two points link back to the vital signs of a healthy crypto portfolio, where concentration and custody are measured side by side, and our portfolio health check computes your effective number of holdings from current values. For the correlation question, see why a diversified portfolio can still move together.
The bottom line
There is no magic number of cryptocurrencies. Correlation sets a floor on how much risk diversification can remove, and in crypto that floor tends to be high, so the first few evenly weighted coins do most of the work. Beyond that, weights, cash and costs matter more than another ticker.
Frequently asked questions
How many cryptocurrencies should a beginner hold?
No single number fits everyone, and the count matters less than people think. In a simple model where coins have 75% volatility and a 0.7 average correlation, the first four or five equal-weight coins deliver most of the risk reduction available, and later ones add very little. How evenly you weight your holdings, and whether you can monitor each one properly, usually matters more than adding another ticker.
Is it better to hold 5 or 20 cryptocurrencies?
In the model used here, five equal coins at 0.7 correlation have 65.4% volatility and twenty have 63.4%, a difference of 2 percentage points. Twenty positions also mean four times as many projects to research, wallets and approvals to secure, and records to keep. Whether that trade is worth it depends on your time and on how differently the extra coins actually behave.
Does diversification work in crypto?
Partly. Spreading money across coins reduces the damage from any single project failing, which is real protection. It does little against market-wide moves, because crypto prices have tended to rise and fall together. The U.S. Financial Stability Oversight Council described this as largely non-diversifiable risk inside the crypto ecosystem. Holding cash-like assets alongside crypto changes total risk in a way extra coins cannot.
What is the effective number of holdings?
It is 1 divided by the sum of your squared portfolio weights, also written 1 ÷ HHI. It tells you how many equal-sized positions would be just as concentrated as your actual portfolio. Four coins at 25% each score exactly 4. Ten coins with one at 55% score far lower, which is why the count on your screen can overstate your diversification.
Sources
- Report on Digital Asset Financial Stability Risks and Regulation (2022) — Financial Stability Oversight Council (U.S. Treasury)
- Beginners' Guide to Asset Allocation, Diversification, and Rebalancing — U.S. SEC — Investor.gov
- Asset Allocation and Diversification — FINRA
This content is for education only and is not financial, investment, tax or legal advice. Crypto assets are volatile and you can lose money. Examples use hypothetical numbers. See our disclaimer and editorial policy.