What is the Best Crypto Product to Invest In?
Ask how much crypto belongs in a portfolio and you get the same answer everywhere. BlackRock says 2%. Fidelity says 2% to 5%. Paul Tudor Jones says 5%. A handful of advisors argue for far more, and the rest of the industry has settled around single digits.
That answer is fine. I have no argument with it.
The problem is that it stops one question too early.
Say you settle on 5% and you want that money spread across Bitcoin, Coinbase, and Strategy. Nobody tells you how to split it. Every article ends right where the actual decision starts.
Here is how I do it.
This closes a three part series. Parts one and two covered when I think Bitcoin bottoms and which vehicle to buy.
Equal dollars is not equal risk
The obvious split is even. A third each. It feels balanced and it looks like diversification.
Run the numbers and it falls apart, because those three things do not move the same amount.
Volatility here means annualized standard deviation, which measures how far a price typically swings around its own average over a year. A 46% reading means moves of roughly that size in either direction are a normal year. Bigger number, wilder ride.
Position | Volatility range, past two years | Midpoint I use |
|---|---|---|
Bitcoin (proxy for IBIT and FBTC) | 40% to 52% | 46% |
COIN | 80% to 102% | 91% |
MSTR | 80% to 134% | 107% |
Source: TradingView historical volatility indicator. 365 days for Bitcoin since it trades every day. 252 days for the stocks, since they only trade when the market is open.
MSTR swings more than twice as hard as Bitcoin does. So a dollar in MSTR carries more than twice the risk of a dollar in Bitcoin.
Split evenly and here is what you actually own. Two thirds of your money sits in the two wildest things on the list. The calmest holding, which is the one you would most want to lean on through a bad stretch, gets the smallest share of your risk. Your outcome now belongs almost entirely to MSTR and COIN.
You have three tickers. You do not have three roughly equal bets.
The fix takes four lines
Bridgewater built a business on this idea decades ago. Size each position so every one contributes about the same amount of risk. Ray Dalio's full version got complicated. The simple version fits on an index card.
Weight each holding by the inverse of its volatility.
Invert each one:
Add them up:
Divide each by that sum:
Which gives you:
Position | Weight |
|---|---|
Bitcoin ETF | 52% |
COIN | 26% |
MSTR | 22% |
In actual dollars
Say your portfolio is $500,000 and you decided on the 5% that everyone recommends. That is $25,000 going into crypto.
Position | Dollars |
|---|---|
IBIT or FBTC | $13,000 |
COIN | $6,500 |
MSTR | $5,500 |
Half the money goes to the calmest asset. The two wild ones split the rest. Every position now contributes roughly the same amount of risk, so no single holding can take out the sleeve on its own.
Compare that to $8,333 in each. You would be carrying almost twice the risk in MSTR that you carry in Bitcoin, without having decided to.
The code
#
# Pull annualized volatility for each ticker from tradingview.com (HV indicator).
# 365 days for crypto, since it trades every day.
# 252 days for stocks, since they follow the market calendar.
#
VOLATILITY = {
"BTC": 0.46,
"COIN": 0.91,
"MSTR": 1.07,
}
def equal_risk_weights(volatilities: dict[str, float]) -> dict[str, float]:
inverse_vols = {ticker: 1.0 / vol for ticker, vol in volatilities.items()}
total = sum(inverse_vols.values())
return {ticker: inv / total for ticker, inv in inverse_vols.items()}
if __name__ == "__main__":
for ticker, weight in equal_risk_weights(VOLATILITY).items():
print(f"{ticker}: {weight:.1%}")
Spreadsheet version, if you would rather not run code: equal risk weight calculator
This works on anything, not just crypto
Nothing in the formula knows what a Bitcoin is. It only cares how much each position moves.
Add a fourth holding like IONQ and every weight adjusts on its own. Using 100% volatility for IONQ, purely as an example, the four weights come out near 42% Bitcoin, 21% COIN, 18% MSTR, and 19% IONQ. Pull the real number before you trade it.
Whatever sits in your high-risk sleeve, this keeps the risk per position roughly level and gives your calmer holdings the room they have earned.
Where this breaks
Volatility looks backward. You are sizing tomorrow's position with yesterday's numbers, and crypto volatility can double inside a month. Recheck the inputs a few times a year and adjust.
The method also ignores correlation, which is how tightly your holdings move together. Bitcoin, COIN, and MSTR all fall on the same bad days. This balances the size of each bet without doing anything about the fact that they are three versions of one bet underneath. Real risk parity handles that. This does not.
And volatility misses risks that have nothing to do with price swings. MSTR carries debt and a premium that can collapse under its own weight. Coinbase carries business and regulatory risk. Neither shows up in a standard deviation.
So treat these weights as a starting point that beats guessing.
Which matters, because things always go wrong in crypto.
Where the series lands
Three questions in order. When to buy. What to buy. How much of each.
The trade itself is still waiting on a signal that has not fired. When Bitcoin breaks its 200-week average and turns back up, you will read about it here first.
Disclaimer: This newsletter is for general educational and informational purposes only and is not financial, investment, legal, tax, or accounting advice. It reflects the author's personal opinions, is not tailored to any individual, and is not a recommendation to buy or sell any security. Investing involves risk, including possible loss of principal, and past performance does not guarantee future results. The author is not a registered investment adviser. Before making any financial decision, consult a qualified professional who can consider your specific situation; you are solely responsible for your own decisions.