Bitcoin's Drawdown Is a Question of Size

Bitcoin fell 53% peak to trough. What that does to a portfolio depends almost entirely on how much of it you own, according to CoinShares’ Research.

Volatility has been the standing objection to bitcoin since it first had a market price, and the objection is fair. Bitcoin has fallen more than 50% ten times in its traded history. It closed 2 August at $63,470, roughly 49% below its 6 October 2025 high of $124,774, having bottomed at $58,551 in June for a peak-to-trough fall of 53%.1

The historical record is the useful frame. Every previous drawdown of that depth resolved in full, by eventually making a new high. No cycle downturn has yet proven permanent.

Whether that pattern holds again is not something an advisor can act on. Position size is.

What a 5% sleeve did

Against a multi-asset baseline returning 4.8% annualised, with 12.3% volatility, a Sharpe ratio of 0.39 and a maximum drawdown of 24.1%, adding a rebalanced 5% bitcoin sleeve produced 8.2% annualised and a Sharpe ratio of 0.67, based on models established from January 2020 to June 2026. Volatility? It was essentially unchanged at 12.2%. Worst-case drawdown widened by 160 basis points, to 25.7%.

That figure is easy to overstate, so read it precisely. The 160 basis points is the gap between two whole-period maximum drawdowns, not the cost of the current episode, and the two portfolios’ worst moments need not be the same event. The claim is narrower than it first appears, and still useful: at a 5% weight, an asset capable of halving moved worst-case portfolio drawdown by less than two percentage points.

Filling the same sleeve differently shifted the trade rather than removing it. A bitcoin and ether split returned 8.7% with a 26.6% maximum drawdown; a top-50 index sleeve returned 9.4% with 26.4%. Diversifying inside the sleeve added return, and slightly more drawdown rather than less.

Content continues below advertisement

Bitcoin is at its 200-week average

Bitcoin is trading at or just below its 200-week moving average, with provider estimates of that level clustering either side of $63,500.2 June marked the sixth time spot has met the average, and on the five completed prior occasions bitcoin was higher a year later.

The sample will not carry much weight. Five episodes across eleven years is not a dataset, and each sat in a different market structure: thin 2015 infrastructure, the March 2020 liquidity shock, the 2022 lending unwind. The average is not confirming anything at present. It is being tested.

Which returns to sizing. A 5% allocation does not require a view on whether the 200-week average holds. It does require a view on bitcoin over a full cycle.

Model figures are hypothetical, gross of fees and taxes, and do not represent the performance of any account or fund.

For more news, information, and strategy, visit the CoinShares Crypto ETF Hub.