DWF Labs: Bitcoin ETF Buyers Follow Trends, Not Set Them

US spot Bitcoin ETF flows have become a popular real-time barometer for market direction, but DWF Labs argues investors should treat daily inflow and outflow data as confirmation rather than a predictive signal.

In a research report released Thursday, DWF reviewed 694 trading days since US Bitcoin ETFs began in January 2024. The firm found that investors tended to move money into the funds after sharp rallies, and withdraw during sizable drawdowns—meaning ETF flow patterns largely reflect what investors are already reacting to, not what Bitcoin is about to do next.

Key takeaways

• DWF Labs reports that during Bitcoin’s sell-offs of at least 15%, net ETF outflows occurred in seven of eight instances.
• Large rallies of the same magnitude saw net inflows in all nine cases studied.
• ETF flows show a limited relationship with Bitcoin’s price changes over the following five trading sessions (correlation of 0.03), suggesting they offer little forecasting value.
• Heaviest withdrawals clustered in two recent declines, with DWF estimating $6.9 billion withdrawn from October 2025 to February 2026 and $8.2 billion during May–June 2026.
• DWF says the data cannot clearly identify whether outflows were driven by specific investor groups, including institutions versus retail, and estimates some flows may be hedged via futures strategies.

Flows track reactions, not forecasts

DWF’s central conclusion is straightforward: ETF flows confirm market moves more reliably than they anticipate them. The report states, “ETF flows confirm moves; they don’t predict them,” arguing the pattern is consistent with investors responding to price action rather than creating it.

To test whether flows could forecast near-term direction, DWF compared daily changes in ETF fund flows to Bitcoin’s performance over subsequent trading windows. The firm found that a day’s ETF flow changes had only a moderate relationship with Bitcoin’s price movement across the five sessions leading up to that day, with a correlation of 0.54. But when the analysis looked forward—measuring the relationship between ETF flows and Bitcoin’s changes over the next five trading sessions—the correlation dropped to 0.03, essentially close to no relationship.

That asymmetry matters for traders and allocators who use flows as a timing tool. Under DWF’s findings, flows may be useful for gauging whether market participants are becoming more risk-averse or more willing to add exposure after volatility hits—but they provide little signal for whether Bitcoin will rise or fall in the days immediately following.

Why outflows don’t reveal who sold

While DWF’s data indicates a consistent pattern of withdrawals during major drawdowns, it does not break down who is behind the selling pressure. The report highlights that the largest net outflows appeared during the two most recent declines: an estimated $6.9 billion withdrawn between October 2025 and February 2026, followed by $8.2 billion during the May–June sell-off period.

However, the figures alone cannot distinguish whether those withdrawals came primarily from institutional investors, retail participants, or a mix of both. In remarks shared with Cointelegraph, a DWF Labs spokesperson said the firm “don’t have a clear split on the exact buyers/sellers of the ETFs,” meaning the flow totals cannot be directly translated into a map of the underlying market constituency.

There is also a structural complication: not all ETF activity necessarily reflects spot exposure in the same way. DWF notes that some investors may buy ETF shares while taking an offsetting position with CME Bitcoin futures—aiming to capture the price spread between the two instruments rather than express a directional view on Bitcoin itself. In the report, DWF estimated that roughly one-fifth of flows could be hedged through such methods, though it could not verify that the same investors both held the ETF position and maintained the futures hedge.

This matters for interpretation. If a portion of ETF flows is driven by hedging and spread strategies, then reductions in net inflows during sell-offs may not correspond cleanly to changes in bullish or bearish conviction among all participants.

DWF’s observations arrive alongside earlier findings from CoinShares. In a June analysis, CoinShares reported that professional investors reduced their Bitcoin ETF exposure by 17% in the first quarter of 2026. According to that CoinShares work, hedge funds and brokerages accounted for about 96% of the reduction, while banks increased their holdings.

Sell-offs appear less frequent, not necessarily smaller

DWF also examined how Bitcoin’s drawdown pattern has evolved since ETF trading began. The report finds that Bitcoin experienced about three sell-offs of 15% or more per year after the ETFs launched. That contrasts with roughly four such events per year in 2022–2023 and eight in 2021.

Alongside the lower frequency, DWF suggests the declines have become more gradual. Even so, the magnitude remains comparable: the median 15%+ sell-off in the ETF era reportedly wiped out 21.3% of value, versus 19.4% in 2022–2023.

The spokesperson also cautioned that ETF flows are not the only driver of volatility trends. They noted that volatility and the frequency of sell-offs were already declining before the funds launched in 2024, implying that changes in market behavior cannot be attributed solely to the introduction of ETFs.

For market participants, this is an important framing. If major drawdowns are becoming less common but still capable of producing similar depth when they occur, then flow-driven narratives may need to be tempered: ETF activity may help describe what investors do during stress, but it does not automatically explain why stress appears—or disappears.

What to watch next

As more traders treat spot Bitcoin ETF flows as a proxy for positioning, DWF’s findings suggest the most reliable use is to monitor how participants react to price—especially during high-volatility periods—rather than expecting flows to call the next swing. The key uncertainty that remains is the investor mix behind net inflows and outflows, and whether hedged strategies continue to shape what appears in daily flow statistics.