Bitcoin closes about $900 below its heaviest long-term holder band with leverage already cleared

Glassnode had earlier called a wider $81,000-$86,000 shelf the main overhead wall, and the heaviest slice of that supply has since narrowed to $84,000-$85,000 ^[3]. Its Week 38 report had the block sitting just below price ^[4]. The CMBI close is now about $887 under the band's lower edge ^[1], after a day down about 1.9% and a week down 3.1% ^[14]. From $83,113, Glassnode's mean MVRV mark near $96,700 is about 16% higher ^[2]. Its True Market Mean near $77,000 is about 7.4% lower ^[3]. The upside mark is a little more than twice as far away as the downside one ^[4].

Coin-denominated open interest is at its lowest since March and nearly 20% below August ^[6], even though the price is still about 35% above August's low near $62,000 ^[7]. Traders have not rebuilt the positions flushed on the run to $87,000, and perpetual futures sit close to neutral ^[8]. According to Crowdfund Insider's reading of the data, that lowers the odds that a reaction at the cluster turns into a liquidation cascade ^[8]. With no crowded short book to squeeze either, a move toward $96,700 would have to be bought in spot.

US spot bitcoin ETFs took in close to $3 billion over 30 days and turned 2026 net flows positive ^[16]. BlackRock's IBIT took most of it, while Grayscale's GBTC still saw periodic redemptions ^[16]. At the CMBI print, $3 billion buys about 36,000 BTC ^[5]. On 22 September alone, holders realized 25,700 BTC of profit, the largest single day of 2026 ^[11]. Valued at $83,113, the day's profit comes to about $2.1 billion ^[6], roughly 70% of the month's ETF intake ^[7]. The comparison sets a profit figure against a purchase figure at one day's price, so it checks scale and nothing finer. CryptoQuant also reports that spot demand has contracted and futures demand growth has slowed sharply ^[12].

The likeliest sellers are short-term traders. Their on-chain unrealized profit margin is 33%, the highest since December 2024, and CryptoQuant's 29 September note says margins that stretched have historically preceded selling ^[10]. The same firm's Bull Score Index stands at 90 out of 100, a reading it labels "extremely bullish", after a close above the 365-day moving average ^[9].

The bullish path has long-term holders sitting on the band while ETF buying absorbs the short-term profit-taking, so price recovers $84,000 and keeps the $96,700 route open ^[5]. A middling one has the band turn into distribution and price drift to CryptoQuant's first support, the 365-day average near $80,000 ^[13], and the firm would still call that consolidation inside a young bull market ^[18]. The bearish one has spot demand keep shrinking until the move reaches $77,000 ^[4], with CryptoQuant's next supports at $71,000 and $67,000 ^[13].

I think the middle path is the likeliest, because price has already closed under the band ^[14] while spot demand contracts ^[12]. The counter-case is the ETF bid. Close to $3 billion in 30 days ^[16] is more than the roughly $2.1 billion of profit realized on the year's biggest day ^[6], and with leverage cleared there is no forced seller to turn a dip into a cascade ^[8]. A daily close back above $85,000 ^[1], with ETF inflows continuing and open interest still low, would prove this view wrong. It would mean spot buyers took the holders' coins without help from leverage. A close under $80,000 would point to the bearish path instead ^[13].