Key Takeaways
- • DWF Labs research, covering 694 trading days since US spot Bitcoin ETFs launched in January 2024, finds ETF flows confirm price moves rather than predict them, with a forward five-session correlation to Bitcoin returns of just 0.03.
- • Net ETF outflows occurred in seven of eight Bitcoin sell-offs of at least 15%, while net inflows followed all nine rallies of comparable magnitude.
- • DWF estimates $6.9 billion was withdrawn from October 2025 to February 2026 and $8.2 billion during the May–June 2026 sell-off, but cannot identify whether institutions or retail drove the outflows.
- • Roughly one-fifth of ETF flows may be hedged via CME Bitcoin futures spread trades, complicating interpretation of inflows and outflows as directional conviction.
- • Bitcoin 15%+ sell-offs have become less frequent, averaging about three per year in the ETF era versus about four in 2022–2023 and eight in 2021, though median drawdown depth is similar at 21.3% versus 19.4%.
Summary
DWF Labs, in a research report released on a Thursday and reported via MENAFN and Crypto Breaking, argues that daily inflow and outflow data for US spot Bitcoin ETFs should be treated as confirmation of market moves rather than a predictive signal. The firm reviewed 694 trading days since US Bitcoin ETFs began trading in January 2024 and found that investors generally added money after sharp rallies and withdrew during sizable drawdowns. DWF's central conclusion is that "ETF flows confirm moves; they don't predict them."
The quantitative findings underpin that view. During Bitcoin sell-offs of at least 15%, net ETF outflows occurred in seven of eight instances, and rallies of the same magnitude saw net inflows in all nine cases studied. A day's flow changes showed a moderate correlation of 0.54 with Bitcoin's price movement over the preceding five sessions, but the correlation fell to 0.03 when measured against the following five sessions, which DWF characterizes as essentially no relationship. For traders and allocators using flows as a timing tool, the report implies flows may gauge risk appetite after volatility but offer little signal on near-term direction.
The report also highlights the limits of attributing flows to specific investor groups. The largest net outflows were an estimated $6.9 billion between October 2025 and February 2026 and $8.2 billion during the May–June 2026 sell-off. DWF said the data cannot clearly separate institutional from retail selling. A DWF Labs spokesperson, speaking to Cointelegraph, said the firm does not have "a clear split on the exact buyers/sellers of the ETFs." DWF also estimates that roughly one-fifth of flows could be hedged by investors who buy ETF shares while holding offsetting CME Bitcoin futures positions to capture the spread, though it could not verify that the same investors maintained both legs. As a result, reductions in net inflows during sell-offs may not map cleanly to shifts in bullish or bearish conviction.
DWF's findings sit alongside a June CoinShares analysis, cited in the report, which found professional investors cut Bitcoin ETF exposure by 17% in the first quarter of 2026, with hedge funds and brokerages accounting for about 96% of the reduction while banks increased holdings. Separately, DWF observes that Bitcoin has experienced about three 15%-plus sell-offs per year since the ETFs launched, compared with roughly four per year in 2022–2023 and eight in 2021. Declines have become more gradual, but the median 15%-plus sell-off in the ETF era reportedly erased 21.3% of value, versus 19.4% in 2022–2023. The DWF spokesperson cautioned that volatility and sell-off frequency were already declining before the 2024 launches, so the ETFs cannot be credited solely for the change. The report concludes that the most reliable use of ETF flow data is to monitor how participants react to price during high-volatility periods, while the investor mix behind flows and the extent of hedged activity remain open questions.
The quantitative findings underpin that view. During Bitcoin sell-offs of at least 15%, net ETF outflows occurred in seven of eight instances, and rallies of the same magnitude saw net inflows in all nine cases studied. A day's flow changes showed a moderate correlation of 0.54 with Bitcoin's price movement over the preceding five sessions, but the correlation fell to 0.03 when measured against the following five sessions, which DWF characterizes as essentially no relationship. For traders and allocators using flows as a timing tool, the report implies flows may gauge risk appetite after volatility but offer little signal on near-term direction.
The report also highlights the limits of attributing flows to specific investor groups. The largest net outflows were an estimated $6.9 billion between October 2025 and February 2026 and $8.2 billion during the May–June 2026 sell-off. DWF said the data cannot clearly separate institutional from retail selling. A DWF Labs spokesperson, speaking to Cointelegraph, said the firm does not have "a clear split on the exact buyers/sellers of the ETFs." DWF also estimates that roughly one-fifth of flows could be hedged by investors who buy ETF shares while holding offsetting CME Bitcoin futures positions to capture the spread, though it could not verify that the same investors maintained both legs. As a result, reductions in net inflows during sell-offs may not map cleanly to shifts in bullish or bearish conviction.
DWF's findings sit alongside a June CoinShares analysis, cited in the report, which found professional investors cut Bitcoin ETF exposure by 17% in the first quarter of 2026, with hedge funds and brokerages accounting for about 96% of the reduction while banks increased holdings. Separately, DWF observes that Bitcoin has experienced about three 15%-plus sell-offs per year since the ETFs launched, compared with roughly four per year in 2022–2023 and eight in 2021. Declines have become more gradual, but the median 15%-plus sell-off in the ETF era reportedly erased 21.3% of value, versus 19.4% in 2022–2023. The DWF spokesperson cautioned that volatility and sell-off frequency were already declining before the 2024 launches, so the ETFs cannot be credited solely for the change. The report concludes that the most reliable use of ETF flow data is to monitor how participants react to price during high-volatility periods, while the investor mix behind flows and the extent of hedged activity remain open questions.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.