Bitcoin fell 14.2% in the second quarter, while institutional Bitcoin ETF holdings rose 7.5% over the same period, climbing from 498,389 to 535,723 BTC equivalent. Fewer institutions drove that increase, with the number reporting Bitcoin positions through 13F filings falling roughly 6.8%, from about 2,000 to around 1,900.
Those figures are Bitcoin Strategy's second-quarter estimate, built from public 13F filings the SEC collects but does not itself aggregate this way.
CoinShares, using its methodology on the prior quarter, counted just 261,000 BTC held by professional 13F filers. It put the professional share of US-traded spot Bitcoin ETF assets at 20.8%, well below Bitcoin Strategy's 44.2% figure.
The gap between those two counts shows how much the headline number depends on which filers and products get counted.
The banks and quants that built Bitcoin positions
JPMorgan's reported ordinary IBIT position rose 25.35% during the quarter, climbing from 8,302,691 shares to 10,407,635 shares, worth roughly $355.7 million at the June 30 filing price.
That is one of the cleaner large-bank comparisons in the entire round, since it measures ordinary shares against ordinary shares, without folding in derivatives or other exposure.
It does not establish that JPMorgan made a $356 million wager on Bitcoin's price, since a consolidated bank 13F can reflect client accounts, hedging, or exposure spread across multiple desks.
Renaissance Technologies posted the sharpest percentage gain among recognizable managers, more than quadrupling its ordinary IBIT position to 1,403,942 shares from roughly 340,000, a 312.92% increase worth about $46.7 million by quarter's end.
The sticky holders may carry the most weight
Several major allocators closed with little to no movement, and that inaction may carry more weight than the bigger percentage swings.
Mubadala Investment Company held its IBIT position flat at 14,721,917 shares, worth about $490.1 million at quarter's end, matching its share count from three months earlier. Harvard Management reported 3,044,612 shares unchanged, and Fortress Investment Group held its 1,325,000 shares steady as well.
The Abu Dhabi Investment Council's position also stayed close to flat, at 8,218,712 shares worth about $273.6 million.
Sticking with a position through a quarter where Bitcoin fell 14.2% behaves very differently from the swings tactical managers made elsewhere in the same filings, even if it falls short of proving permanent conviction.
What the tactical Bitcoin reducers cut
Brevan Howard's ordinary IBIT position fell 70.36% during the quarter, from 24,304,788 shares to 7,205,004. Read alone, that looks like one of the clearest bearish reads in the entire round.
The same filing complicates that read. Brevan Howard also reported IBIT call options tied to roughly 7.23 million underlying shares and put options tied to about 5.27 million, which supports a narrower claim than a simple bearish trade.
Brevan Howard cut most of its ordinary IBIT shares, as the 13F shows, though whether it exited Bitcoin altogether is a separate question the filing alone cannot answer.
Citadel's filing raises the same problem. Its ordinary IBIT position dropped 59.66% to 514,614 shares, dwarfed by the options book beside it: calls tied to about 24.65 million underlying shares and puts tied to roughly 17.86 million.
Characterizing that filing as a simple bearish Bitcoin trade would mean ignoring most of the position.
Macquarie's ordinary IBIT position fell a similar 61.78%, to 1,581,934 shares worth about $52.7 million. Macquarie also functions as an authorized participant able to create and redeem IBIT shares, the same plumbing role that complicates reading Jane Street's filing.
Why Jane Street's billion-dollar position reads differently
Jane Street's reported IBIT position rebounded from 5,872,212 shares in the first quarter to 24,878,191 shares in the second, worth about $828.2 million at quarter's end. Taken at face value, that would be one of the largest directional Bitcoin bets in the entire filing season.
Jane Street is also listed among BlackRock's authorized participants for IBIT, the firms responsible for creating and redeeming ETF shares to keep the fund's price in line with the underlying Bitcoin it holds.
Quarter-end inventory for a firm in that role can reflect market-making and hedging as much as any directional view. Jane Street's own filing shows options tied to roughly 34 million and 34.8 million underlying shares on the call and put sides.
That combination reads far more like ETF plumbing than a bullish wager.
Every 13F treats an ETF share the same way, whether it sits with a sovereign wealth fund that has held it untouched for two straight quarters or a market maker cycling inventory daily.
Those are meaningfully different owners, and lumping them into one institutional total obscures the difference between demand that tends to persist through a drawdown and demand that tends to trade through one.
What determines whether the sticky base grows
The bull case has the second quarter's quiet holders staying quiet, or adding, through the third quarter, while banks and wealth platforms keep building alongside them and the number of institutional filers stabilizes.
If sovereign, endowment, and advisory-platform ownership keeps expanding while spot ETF inflows turn consistently positive, Bitcoin's institutional float starts looking genuinely less reflexive during the next downturn.
The bear case has the headline institutional number staying strong for reasons that have little to do with durable demand, driven mainly by trading firms, options books, and ETF plumbing that dominate every quarter-end snapshot.
Sticky holders remain present without expanding their positions, and the next real drawdown shows how much of the second quarter's institutional total was always closer to inventory than conviction.
Bitcoin's second-quarter filings show that the institutional ownership base is already splitting into holders built to stay and holders built to trade. Only one of those groups says much about what happens the next time Bitcoin falls 14% in a quarter.
