Bitcoin ETFs belong in a diversified ETF portfolio only at a size that survives their drawdowns. In 2026 the iShares Bitcoin Trust ETF (IBIT) fell 40.0% from its January 14 closing high of $55.44 to a June 30 low of $33.29, then recovered to $47.88 by September 23, 2026, still 3.6% below its year end 2025 close [5]. Bitcoin itself traded at $84,383 on September 23, 2026, 44.1% above its June 30 low of $58,559 and about 33% below its 52 week high of $126,198 [4]. Daily returns on IBIT had a correlation of 0.45 with the iShares Russell 2000 ETF (IWM) and 0.29 with the Vanguard High Dividend Yield ETF (VYM) in 2026 through September 23, with annualized volatility of 46.8% against 18.5% for IWM [5]. Those three numbers, a 40% drawdown, a moderate correlation and volatility near 50%, are what sizing should be built on.
What bitcoin ETFs hold and what they cost
Spot bitcoin ETFs are grantor trusts that hold bitcoin with a custodian and issue shares that track its price less a sponsor fee. IBIT states that it is not an investment company registered under the Investment Company Act of 1940 and not a commodity pool under the Commodity Exchange Act, so it lacks the diversification and governance rules that apply to ordinary ETFs [1]. Its sponsor fee is 0.25%, the same rate Fidelity charges on the Fidelity Wise Origin Bitcoin Fund (FBTC) [1][3]. The iShares Ethereum Trust ETF (ETHA), launched June 24, 2024, applies the same 0.25% fee to ether [2]. There is no yield: the trusts hold a non income producing asset, so the fee is a certain annual drag and the return is entirely price change [1].
The 2026 record shows what that exposure means. IBIT's total return at NAV was a loss of 32.97% for the first half of 2026 and a loss of 45.62% for the year to June 30, 2026, although the fund still showed a 33.05% annualized return since its January 5, 2024 launch [1]. ETHA lost 46.90% in the first half and 37.24% over the year to June 30 [2]. Bitcoin's own path, tracked on DataPorium's crypto page, ran from $96,929 on January 14, 2026 down to $58,559 on June 30 and back to $84,383 on September 23 [4].
Correlation: how bitcoin ETFs moved with stocks in 2026
Diversification depends on correlation, and bitcoin's has been moderate rather than low. Using DataPorium daily closes from January 2 to September 23, 2026, the correlation of IBIT's daily returns was 0.45 with IWM, 0.30 with the iShares Semiconductor ETF (SOXX), 0.29 with VYM and slightly negative, at minus 0.10, with the Utilities Select Sector SPDR Fund (XLU) from April onward [5]. By comparison, VYM and IWM had a correlation of 0.74 with each other, and SOXX and IWM 0.67 [5]. Bitcoin therefore behaved like a high beta risk asset that shares some of the stock market's direction but not most of its day to day moves.
Volatility and drawdown in 2026
| ETF (January 2 to September 23, 2026) | Annualized daily volatility | Largest peak to trough decline | Correlation with IWM |
|---|---|---|---|
| iShares Bitcoin Trust ETF (IBIT) | 46.8% | -40.0% (Jan 14 to Jun 30) | 0.45 |
| iShares Semiconductor ETF (SOXX) | 49.9% | -29.0% (Jun 22 to Jul 29) | 0.67 |
| iShares Russell 2000 ETF (IWM) | 18.5% | -11.2% (Jan 22 to Mar 30) | 1.00 |
| Vanguard High Dividend Yield ETF (VYM) | 10.3% | -7.2% (Feb 11 to Mar 20) | 0.74 |
Source: DataPorium daily prices [5]. Bitcoin's volatility was in the same range as the most volatile equity sector fund of the year, and its drawdown was deeper, but its correlation with small caps was lower than that of either SOXX or VYM. A small position can therefore reduce the share of portfolio risk that comes from the stock market, provided the position is small enough that its own swings do not dominate.
Sizing a bitcoin ETF position: the arithmetic
The simplest sizing test is to apply the 2026 drawdown to the position and ask whether the portfolio impact is tolerable:
- 1% allocation: a 40% decline costs 0.4% of the portfolio, less than a typical bad week for a broad equity fund [5].
- 2% allocation: a 40% decline costs 0.8%; a full 100% loss, which the trust structure does not rule out, costs 2%.
- 5% allocation: a 40% decline costs 2.0% of the portfolio, and with 46.8% volatility the position contributes a meaningful share of total risk [5].
- 10% allocation: a 40% decline costs 4.0%, more than the entire 2026 drawdown of a diversified equity fund such as VYM (7.2%) would have cost at a 50% weight [5].
Rebalancing matters as much as the starting weight. A 2% position that doubles becomes 4%, and one that halves becomes 1%; trimming back to target after large moves forces the investor to sell high and buy low, which is the only reliable way to harvest an asset with 40% swings and no cash flow. Rebalancing does create taxable gains in a brokerage account, so some investors hold the position in a tax deferred account.
The counterpoint deserves space. Bitcoin has produced a 33.05% annualized return for IBIT holders since January 2024 despite the 2026 decline, and the fixed supply argument that attracts investors is not affected by a bad year [1]. The case against a larger weight is not that the asset is bad but that its risk is unpriced in the usual sense: there is no earnings stream, no yield and no regulatory diversification, so the only defense is size. Investors may consider a weight between 1% and 5%, set by how much of the 2026 drawdown they could have watched without selling.
At a 1% to 5% weight a bitcoin ETF adds a moderately correlated asset to a portfolio; above that, its 47% volatility and 40% drawdowns start to set the portfolio's risk.
Key takeaways
- IBIT fell 40.0% from January 14 to June 30, 2026 and was still 3.6% below its year end 2025 price on September 23 [5].
- Bitcoin recovered 44.1% from its June 30 low to $84,383 on September 23, 2026, about 33% below its 52 week high [4].
- IBIT's 2026 correlation with small cap stocks was 0.45 and with dividend stocks 0.29; its volatility was 46.8% [5].
- IBIT, FBTC and ETHA each charge 0.25%; the trusts hold no income producing assets and are not 1940 Act funds [1][2][3].
- A 1% to 5% allocation limits the portfolio cost of a 40% drawdown to 0.4% to 2.0%; regular rebalancing captures the swings [5].
Frequently asked questions
How much bitcoin should be in a diversified portfolio?
The 2026 data suggest a size where a 40% decline is tolerable: at 1% to 5% of the portfolio that loss would cost 0.4% to 2.0% of total value, while a 10% weight would cost 4.0% [5].
Is bitcoin correlated with the stock market in 2026?
Moderately. IBIT's daily returns had a 0.45 correlation with the Russell 2000 ETF (IWM) and 0.29 with the dividend ETF VYM from January to September 23, 2026, compared with 0.74 between VYM and IWM [5].
What does a bitcoin ETF cost?
IBIT charges a 0.25% sponsor fee, as does Fidelity's FBTC and the ether trust ETHA; the trusts pay no income, so the fee is the full annual drag [1][2][3].
How much did bitcoin ETFs lose in 2026?
IBIT's NAV total return was minus 32.97% for the first half of 2026 and minus 45.62% for the year to June 30, 2026; by September 23 the fund had recovered to 3.6% below its year end 2025 price [1][5].
