The sleep test is the only sizing tool I’ve found that survives contact with a real bear market. I wrote about it in how much bitcoin you should own: the sleep test — the idea that your allocation should be small enough that you sleep normally while it drops 70% and boring enough that you don’t check it hourly while it doubles. This post is the practical follow-up: the common sizing frameworks, what each implies in dollars and drawdowns, rebalancing rules that keep a volatile asset from quietly eating your portfolio, and the custody question that decides whether you actually own what you bought.
Standard framing: I’m not a financial advisor, bitcoin is genuinely speculative, and nothing here is a recommendation to buy it. This is a map of how people who do hold it tend to size the position sanely.
The Range Most Frameworks Land In: 1–5%
Walk through what each level means for a hypothetical $200,000 long-term portfolio (illustrative numbers, not anyone’s real balance):
| Allocation | Dollar amount | If bitcoin drops 80% | Portfolio impact |
|---|---|---|---|
| 1% | $2,000 | −$1,600 | −0.8% (a bad week, forgotten in a year) |
| 3% | $6,000 | −$4,800 | −2.4% (annoying, survivable) |
| 5% | $10,000 | −$8,000 | −4.0% (a real hit; needs conviction) |
| 10% | $20,000 | −$16,000 | −8.0% (a bet, not a satellite) |
The honest tradeoff runs the other way too. If bitcoin quadruples, a 3% position adds roughly 9% to your portfolio. A 10% position adds 30%. The position size you choose is really a decision about which regret you’d rather live with: “I only had a little when it ran” or “I had too much when it collapsed.” There is no allocation that removes both regrets. Pick the one whose pain you’d rate as tolerable.
Why an 80% Drawdown Is the Planning Baseline
This isn’t hyperbole — bitcoin has declined on the order of 80% from peak to trough multiple times in its public history (2011, 2013–15, 2017–18, 2021–22 all saw drawdowns in that neighborhood). Whatever you believe about its future, sizing that assumes smaller drawdowns is sizing that hasn’t read the history. The 1–5% band exists precisely because a satellite position sized there can’t force bad decisions: even a full collapse is a portfolio event, not a life event.
The corollary matters more than the sizing rule: no money you need within a few years belongs in bitcoin. Not the emergency fund, not the house down payment, not next year’s tuition. The asset’s volatility makes it categorically wrong for near-term money, whatever you believe about its long-term case.
Rebalancing: The Rule That Does the Real Work
Here’s the mechanical problem nobody plans for: a small bitcoin allocation doesn’t stay small. Say you set 3% of a $200,000 portfolio — $6,000. Bitcoin doubles. Now you have $12,000 in a $206,000 portfolio — 5.8%. Bitcoin does it again and you’re at 10%+ without ever “deciding” to increase your risk. Volatile assets drift upward when they win, and your carefully chosen 3% quietly becomes the bet you swore you’d never make.
Two clean ways to handle it:
- Band rebalancing: set a range, like 2–4%. When bitcoin’s share crosses a boundary, trade it back to target. Above 4%: trim and move proceeds into your boring portfolio. Below 2%: top up (only if the original thesis still holds). This turns volatility into a mechanical advantage — you are systematically selling high and buying low without ever making a market call.
- Calendar rebalancing: once or twice a year, whatever the price did, reset to target. Simpler, slightly less responsive, and much easier to actually follow.
Worked example of the trim: target 3%, bitcoin doubles, share drifts to ~5.8%. Trimming back to 3% moves roughly $5,800 out of bitcoin and into the rest of the portfolio. You’ve just crystallized the win — the move most bitcoin holders never make, because holding through the top feels disloyal to the thesis. Rebalancing rules exist to make that decision for you, in advance, when your judgment is calm.
The Buy Discipline Question
Beyond sizing, there’s the how of acquiring the position. Two approaches I consider defensible: a one-time purchase of the full target size (clean, done, no timing decisions), or dollar-cost averaging into the position over months (slightly worse expected math if the market rises, better sleep if it doesn’t — the same tradeoff from lump-sum-vs-DCA that applies to every volatile asset). What I’d avoid: size-creep, where every dip tempts you to “just add a little” until the 3% position is 12%. Write the target size down. When the position hits it, you’re done buying.
Custody: The Part That Decides Whether You Own It
Sizing correctly means nothing if the coins can evaporate. Bitcoin is bearer-style property: whoever controls the private keys controls it. That gives you two fundamentally different custody models, and most holders use a hybrid:
- Exchange custody: the platform holds the keys. Convenient, simple, and it carries real history — exchanges have failed outright (Mt. Gox in 2014, FTX in 2022 are the famous ones), and in those cases account holders became unsecured creditors of a bankrupt company. If you leave coins on an exchange, you’re trusting an institution, and the trust should be sized like any counterparty risk.
- Self-custody: you hold the keys, typically on a hardware wallet. No platform can fail and take your coins with it — but now you’re the institution, and lost seeds and botched transfers are the main way people lose coins. There’s a learning curve, and it deserves one careful weekend, not a rushed evening.
My honest middle ground, which is common among long-term holders: keep a small working balance somewhere convenient if you actively trade, and move the long-term core to self-custody once it’s large enough that an exchange failure would actually hurt. I cover the mechanics and the tradeoffs more fully in where crypto actually lives: custody explained, and the hardware-wallet basics are step-by-step in hardware wallets for beginners.
A Sizing Worksheet You Can Actually Fill Out
- Portfolio value: your long-term invested assets, not your net worth including the house.
- Write your candidate allocation: somewhere in 1–5% if you’re conservative-to-curious; above that, be honest that it’s a conviction bet.
- The drawdown test: multiply allocation × 0.80. That’s the dollar loss you’re pre-accepting. Say it out loud. If the number makes your stomach drop, cut the allocation until it doesn’t.
- The upside test: multiply allocation × 3. If bitcoin quadruples, that’s the gain — sanity-check that it’s meaningful enough to justify holding a volatile asset at all, otherwise why bother.
- Set the rebalance band (say ±1 point around target) and the calendar (annually). Write both down.
- Decide custody before the first buy, not after the balance gets scary to think about.
What I Actually Believe
I hold some bitcoin, I find the long-term case interesting, and I hold it small — sized so that a multi-year bear market would be a story I tell at dinner, not a problem I solve by selling. The most honest thing I can tell you: nobody knows where this goes. The 1–5% framework isn’t a forecast; it’s a way to participate in the upside case without betting your financial freedom on an unresolved question. If the position is small enough to lose and meaningful enough to matter, you’ve captured the asymmetry that makes this asset interesting — and made yourself immune to the two failure modes: being talked out of a sane plan, and being talked into a reckless one.
One forward step: run the sleep test on your own number tonight. Pick your candidate allocation, multiply by 0.80, and imagine reading that loss on a red Monday morning with no bottom in sight. If you’d still be holding in year three of a bear market, the size is right. If you wouldn’t, it’s too big — and no amount of conviction changes that math.
