Context and references.

The spot Bitcoin ETF complex completed its second full year of trading in January 2026, with cumulative net flows above $50 billion across the eleven funds tracked by Farside Investors. BlackRock IBIT, Fidelity FBTC, and the Grayscale converted GBTC dominate flow share; the remaining eight funds added about $7.5 billion, roughly 14% of the cumulative net total (Farside daily data to 31 January 2026: IBIT about +$62.0B, FBTC +$11.3B, GBTC -$25.7B, total about +$55.0B).

May 2026 was a distribution month. Farside's daily totals summed to about -$2.4 billion over 20 trading days, with 14 of the 20 days negative; the six sessions from 12 to 19 May alone lost about $2.0 billion, and IBIT accounted for roughly $1.08 billion of that. SoSoValue and Farside Investors both publish the daily aggregate.

ETF flow and BTC direction tend to move in the same week. Matching Farside daily totals against Binance BTCUSDT weekly candles (Monday to Sunday, UTC) from January 2024 to 17 May 2026: of the 69 days with $500M+ net inflow, 53 fell in a week BTC closed higher (about 77%); of the 53 days with $300M+ net outflow, 38 fell in a week BTC closed lower (about 72%). Same-week overlap says nothing about which one leads.

Bear-case considerations: the ETF flow regime can shift quickly. After three straight weeks of $1.6-3.4 billion inflows in November 2024, the week of 25 November ended with a small net outflow (about -$135M) while BTC sat just under its then-record $99,588 (Binance, 22 November).

Where the numbers on this page come from.

Every figure above is a net number — creations minus redemptions, summed across the funds and reported in dollars, not in BTC. Farside Investors publishes it as a daily table with one column per fund; the totals column is what the monthly and weekly sums above refer to. SoSoValue runs the same count independently. Neither is a live feed. Both post after the US close, so a Tuesday number is not readable until Wednesday, and any week containing a market holiday will look short by a day unless you check the column headers.

Check both on the same evening and you will sometimes find they disagree on a given day — one showing a small inflow where the other shows a small outflow. That gap is timing and late issuer revisions, not a scandal, and it is why single-day prints near zero are worth nothing. Compare weekly sums instead; at that horizon the two converge. The $500M threshold quoted earlier sits deliberately far outside that noise band, which is the only reason it holds up as a rule of thumb at all.

The thing neither source publishes is who was on the other side. A net inflow cannot tell you whether the buyer is an allocator building a first position or a basis trader hedging a short CME future — and those two have opposite implications for what happens next. That gap is the real reason to treat the 77% same-week hit rate above as a historical tendency rather than a mechanism you can lean on.

Crypto assets are volatile and not suitable for every investor. This page is editorial analysis, not financial advice.