Position Sizing for an Asset That Falls 70 Percent
Most sizing advice assumes normal volatility. Here is how to think about an asset where a decline of two thirds is an ordinary event.
Bitcoin has fallen more than seventy percent from a peak on several occasions. Not as a crisis event, but as a normal feature of its history. Most alternatives have fallen further and not recovered.
Any sizing approach that treats a seventy percent decline as a tail risk is mis-specified. It is a base case.
Start from the drawdown, not the return
The usual approach is to start from an expected return and work out how much to allocate. For this asset class that is backwards, because the expected return is unknowable and the drawdown is well documented.
Better question: what is the largest amount whose loss would not change any decision I make about my life?
Then size so that a seventy percent decline in the position stays within that amount. If you can afford to lose ten thousand, the position tolerates roughly fourteen thousand, and that is the ceiling rather than the target.
The three numbers worth writing down
The ceiling. The maximum share of investable assets this position may occupy. Mine is expressed as a percentage, set in a flat month, and written down.
The trim threshold. The level above the ceiling at which I sell back to target. Appreciation pushes you past the ceiling without any decision on your part, and without a threshold you simply drift.
The floor. The level below which I stop rebalancing into it. Adding to a falling position is defensible up to a point, and past that point it is how a bad allocation becomes a catastrophic one.
Why the ceiling is not the target
Two people with identical ceilings behave differently if one sits at the limit and the other at half of it. The one at the limit has no capacity to add during a decline, which is the only moment when adding is attractive.
I run at roughly half my ceiling by default. The gap is not idle. It is the thing that lets me act when everyone else is forced to.
Rebalancing schedules beat rebalancing feelings
Quarterly, on a date, regardless of the chart.
The reason is not that quarterly is optimal. It is that a schedule cannot be talked out of, and a feeling can. Every discretionary rebalance I have made has been worse than the mechanical one would have been, and I have kept the records to check.
What this does to returns
It lowers them, in the good scenarios. A ceiling means selling into strength, and in a cycle that keeps going up, selling into strength is expensive.
That is the correct trade. The strategy is not designed to maximise the outcome in the best case. It is designed to produce an outcome you will still be participating in after the worst case, which is the only way the long-run numbers ever reach you.
The practical setup
Long-term holdings sit in cold storage. A working balance stays on the exchange I have used since 2021 so that a scheduled trim does not turn into a two-day project involving a device in a drawer.
The size of that working balance is itself a decision: large enough to rebalance without friction, small enough that losing it entirely would be an annoyance rather than an event.
The rule underneath all of this
Never hold a position size you did not explicitly choose.
Almost every serious loss I have seen, in myself and in people I know, traces back to holding an amount that arrived through appreciation rather than through a decision. The market gives you that position quietly. Taking it back is the entire job.
This is a personal account of holding through market cycles. It describes what one person did and why. It is not a recommendation, and past cycles do not predict future ones.