A 10% Crypto Position and a 10% Bond Position Aren't the Same 10% of Risk. Here's What Risk Parity Fixes
5 min read
Loading...
5 min read
Say your portfolio holds 80% in stocks, 10% in bonds and 10% in crypto. Two of those slices are exactly the same size on paper: 10% bonds, 10% crypto. But if either one has a bad week, they will not hurt you by anywhere near the same amount. One of them barely moves the needle. The other can swing your whole portfolio's return on its own.
That gap between dollar weight and actual risk is what risk parity is built to correct.
A position's dollar size tells you how much money is riding on it. It tells you almost nothing about how much that money can swing.
Stocks, bonds and crypto move by very different amounts in a normal year. Using rolling one-year volatility figures from PortfoliosLab (retrieved 2026-09-04): the State Street SPDR S&P 500 ETF (SPY) carried 12.89% annualized volatility, the iShares Core U.S. Aggregate Bond ETF (AGG) carried 3.79%, and Bitcoin (BTC-USD) carried 35.76%, roughly nine times the bond figure and nearly three times the stock figure.
Put a 10% slice of your money into the bond fund and a 10% slice into Bitcoin, and you have put the same dollar amount at risk but a very different amount of actual, expected price movement. Equal dollar weight is not equal risk. It only looks that way if you never check the volatility behind each line.
Risk parity is a portfolio construction approach that sizes each position by its contribution to total portfolio risk, not by how many dollars sit in it. A low-volatility asset gets a bigger dollar allocation. A high-volatility asset gets a smaller one, so that no single holding quietly dominates how the whole portfolio actually behaves.
The best-known version is Ray Dalio's All Weather portfolio, built at Bridgewater Associates starting in 1996 for Dalio's own family trust before it became a widely followed template. The classic mix runs roughly 30% stocks, 55% long and intermediate Treasury bonds, and 15% split between gold and broader commodities, a dollar split that looks bond-heavy on paper but is designed so stocks, bonds and commodities each contribute a comparable share of the portfolio's actual risk.
A simplified way to estimate how much risk a position contributes, ignoring how the assets move relative to each other, is to multiply each position's dollar weight by its own volatility, then see what share of the total that position represents.
(Weight × Volatility) ÷ Sum of (Weight × Volatility) across all positions
This is a simplification. It leaves out correlation, how much two assets tend to move together or apart, which real risk parity models account for using a full covariance matrix across every holding, not just each position's own volatility in isolation. Correlation can push the true combined risk higher or lower than this simplified version suggests. But the simplified version is enough to show the core problem with equal dollar weighting, which is what the worked example below does.
Take the 80% stocks, 10% bonds, 10% crypto portfolio from the opening, using the same three volatility figures above.
| Dollar weight | Volatility | Weight × volatility | Share of total risk | |
|---|---|---|---|---|
| Stocks (SPY) | 80% | 12.89% | 10.31 | ~72.3% |
| Bonds (AGG) | 10% | 3.79% | 0.38 | ~2.7% |
| Crypto (BTC-USD) | 10% | 35.76% | 3.58 | ~25.1% |
A true risk parity version of this same three-asset mix would shrink the crypto weight, or grow the bond weight, until each position's share of risk looked closer to even, not until each position's share of dollars did.
Concentration risk is about a single position taking up an outsized share of your dollars, the classic case being one stock at 40% of a portfolio. Risk parity is a different question entirely: even a modest, unremarkable-looking 10% position can carry an outsized share of your risk if its own volatility is high enough, with no concentration problem in the dollar-weight sense at all. A portfolio can pass a concentration check and still be quietly dominated by its most volatile small position.
Two things the simplified formula above leaves out matter in practice. First, correlation: if two assets tend to fall together, their combined risk is worse than either one's volatility alone suggests; if they tend to move independently or oppositely, the combined risk can be lower than a naive sum implies. Second, institutional risk parity funds often use leverage on their lowest-volatility holdings, typically government bonds, to bring a naturally low-risk asset's dollar weight up high enough to actually match the other holdings' risk contribution, which introduces its own borrowing and margin risk that a simple, unleveraged retail portfolio does not carry.
You do not need to build the covariance matrix yourself. NetWort's Portfolio Health section on your dashboard computes this from your actual holdings: a dual-ring chart shows each position's real dollar weight next to its real share of total portfolio risk, built from a full covariance matrix across everything you hold, not the simplified single-asset version used above. If a position's risk share sits well above its weight share, that is the same gap this article just walked through, showing up in your own numbers instead of a hypothetical one.