Bitcoin Stuck Between $62,662 and $65,000 as Long-Term Holders Shed 356,000 BTC

Bitcoin (BTC-USD) traded $64,401.56 at 4:54 a.m. ET Wednesday and $64,339.33 by 6:30 a.m., up $203.85 from the same hour Tuesday, before pushing to $64,821.82 by mid-morning for a 1.03% gain. The prior 24-hour band ran $63,246 to $64,516 on $21.35 billion of volume, with the CoinDesk-tracked 24-hour figure at $7.68 billion. Market capitalization sits at $1.33 trillion against a fully diluted valuation of $1.35 trillion, on a circulating supply of 20,071,518 coins — 96% of the 21 million cap.

The number that governs this tape is not on any crypto chart. The 30-year U.S. Treasury yield hit 5.338%, a multi-decade high, before retreating slightly Wednesday morning. The 10-year eased to 4.70%, down 0.21% on the session. Those two prints are why Bitcoin bounced off $63,246 overnight rather than breaking it, and they are why the entire $62,662–$65,000 band has held for three weeks.

The overnight cross-asset picture was ugly everywhere except crypto. South Korea's KOSPI sank 5.8%. Japan's Nikkei 225 dropped 3.16%. SK hynix fell a further 10% and remains roughly 50% below its all-time high despite announcing a 40 trillion won share buyback — $29 billion — and lifting its shareholder return target above 50% of free cash flow. Bitcoin absorbed all of that and finished up on the day. Solana led the majors at $77.41, up 1.52%. Ethereum traded $1,920.67, up 1.02%. XRP held $1.00. The CoinDesk 20 gained 0.79% to $1,751.57. Bitcoin dominance holds near 60%.

Oil moved the other way. WTI and Brent both rose more than 0.5% Wednesday, reaching $85 and $92 respectively. Gold ripped 1.70% to $4,495.60, up more than 10% on the month. Long-term German and French yields kept climbing with no respite.

The setup into the 2:00 p.m. ET release of the July FOMC minutes is a coiled one: 30-day realized volatility has collapsed to 27.2% annualized against a long-run average near 80%, and price has compressed into a band barely 3.7% wide. Something breaks that band this week. The direction depends on the term premium, not on anything native to the network.

The Range Is Two Numbers Wide and Both Are Known

The floor is $62,662. That was the early-August low, and price came within $121 of retesting it on August 14 when the session low printed $62,783. The July range low sits at the same shelf. Every attempt to break beneath it since early July has failed, which makes it the single most defended level on the chart and the obvious liquidation magnet if it goes.

The ceiling is $65,000. Sellers have defended that region on every approach for three weeks. Bitcoin tagged $65,000 intraday on August 18 as equities rebounded from two-week lows on shifting U.S.-Iran rhetoric, and the move was immediately labeled a low-volume liquidity trap rather than genuine demand. The rejection came fast. The prior spike to $64,500 got the same treatment.

Between those two poles the intermediate markers are $63,174 on the downside and $64,557 on the upside. Losing $63,174 puts $62,783 and then $62,662 into play in a single session. Clearing $64,557 opens the $65,000 test, and above that the structure changes character entirely: $65,500, then $66,500, then the far larger $70,000 zone that has not been touched since the failed breakout attempt collapsed.

Price closed August 11 at $63,549, essentially flat month over month at -0.3%. It closed the week ended August 14 at $63,395, down 0.02% on the day. It sits near $64,400 today. That is a 1,000-point drift over eight sessions — the definition of a market that has run out of both sellers and buyers at these levels.

The mid-August breakout attempt has been fully unwound, which returns price to the range that has contained it since early July. Volume tells the same story: $7.68 billion over 24 hours on the CoinDesk tape is thin for a $1.33 trillion asset, and thin markets do not sustain directional moves.

The one asymmetry favors the downside on speed rather than probability. A break of $62,662 has $60,000 as the next visible shelf, a 4.2% air pocket. A break of $65,000 runs into the 100-day EMA within 2.7%. Down moves have further to travel before they find structure.

The Moving Average Stack Is the Entire Bear Case

Bitcoin trades below every meaningful exponential moving average, and the spacing between them describes exactly how much repair work is required before the medium-term trend flips.

The 20-day EMA sits at $64,147, recently $64,010. Price is fractionally above it. That is the only moving average currently under the market, and it is the least significant of the four. The 50-day EMA sits at $64,467, essentially on top of spot, which is why the $64,400–$64,500 zone has become a magnet. The 100-day EMA sits at $66,735, recently $66,604 — 3.6% above the current print. The 200-day EMA sits at $71,925, a full 11.7% above spot.

That configuration — 20 below 50 below 100 below 200, with price wedged between the 20 and 50 — is a corrective structure by any technical definition. The medium-term trend does not turn until the 100-day is reclaimed, and the long-term trend does not turn until the 200-day is reclaimed. At current velocity, the 200-day would require a move that Bitcoin has not managed in a single leg since the breakdown began.

Momentum confirms the compression rather than resolving it. The daily MACD has crossed back down, reading 47 against its 75 signal line, turning the histogram to -27. RSI has fallen to 46.31, below both the 50 midline and its own 49.81 signal. Neither reading is oversold. Neither is overbought. Both sit in the dead zone where momentum indicators provide no edge and price is being set by flow rather than by positioning.

The practical sequence for anyone building a level ladder: reclaim $64,147 and hold it, then take $64,467, then $64,557, then $65,000. Only after $65,000 does $66,735 become a realistic destination rather than an aspiration. Each of those four steps sits within 1% of the next, which means the reclaim can happen fast if it happens at all — a single 3% session covers the entire ladder.

Failure at any rung sends price back to the $63,174 shelf, and the round trip through this range has now happened enough times that the compression itself is the signal.

ETF Flows Turned Positive Again — Against a $4.5 Billion Deficit

U.S. spot Bitcoin ETFs pulled in $189 million in net inflows on August 18. Ethereum funds added $71.468 million over the same stretch, driven largely by BlackRock's ETHA. A single positive session does not reverse anything, but the context around that number is what matters.

The August sequence has been violently two-sided. The week of August 3 through August 7 delivered $853.54 million in net inflows, the strongest weekly haul since mid-April. BlackRock's iShares Bitcoin Trust (IBIT) captured $693 million of that, roughly 81% of category flow. Between August 3 and August 5 alone the complex absorbed about $626 million, with IBIT taking approximately $478 million.

Then it reversed. August 10 posted a $144.7 million outflow. August 11 managed $4.9 million of inflow. August 12 lost $61.2 million, August 13 lost $131.1 million, and August 14 lost $57.6 million — a third consecutive outflow session, capping a week that removed roughly $390 million. Total net assets across the funds stand at $76.6 billion, with $51.8 billion of cumulative net creations since the January 11, 2024 launch.

The structural deterioration is measurable. Across 666 trading sessions from launch through August 14, net flows were negative on 266 of them. The annual breakdown moves in one direction: 31% of sessions negative in 2024, 40% in 2025, and 54% so far in 2026. More than half of all sessions this year have seen the complex shrink.

The record marks belong to this year as well. The longest outflow streak ran 13 sessions from May 15 to June 3, shedding $4.37 billion. The largest 30-day net outflow on record, roughly $6.35 to $6.4 billion, was set in the window beginning mid-May. Year to date, the category sits approximately $4.5 billion in the red despite the August rebound.

One product ceased operations entirely: Hashdex's fund stopped trading on NYSE Arca on August 17, leaving the survivors to absorb whatever demand remains.

The $90,200 Cost Basis Problem Sitting Inside IBIT

The mechanical issue with the ETF complex is not flow direction. It is entry price. Using fund-level inflow timing, the average cost basis for U.S. spot Bitcoin ETF holdings sits around $90,200 per coin. At $64,400, the aggregate holder of these products is underwater by roughly 28.6%.

That number explains behavior that otherwise looks irrational. Every rally toward $65,000 meets supply not because anyone is taking profit but because holders sitting 25% to 30% down use strength to reduce. Every approach to the range high is a redemption opportunity for someone who bought at $90,000 and has been waiting for a place to exit that does not require selling into a vacuum.

The feedback loop cuts both ways. When redemptions of size force sponsors to sell spot Bitcoin into a falling tape, lower prices trigger more risk-management selling from ETF holders, and IBIT sits in the middle as both the largest single holder and the largest source of incremental supply. That mechanism produced the $6.4 billion 30-day drawdown in May and June. It has not been dismantled — it has simply gone quiet because price stopped falling.

IBIT's scale is the reason its direction determines the category. The product became the fastest ETF in history to cross roughly $70 billion in assets and has drawn approximately $60 billion of cumulative net creations in about two and a half years. That remains one of the most successful launches the industry has produced. It also means that when BlackRock's flow turns, nothing else in the complex is large enough to offset it.

The asymmetry in flow shape is the detail that matters for anyone modeling supply. Outflows arrive in concentrated bursts — a single session in mid-July produced a $424.7 million redemption, the largest in three months. Inflows arrive in a drip. One bad day has repeatedly erased close to half of what a full week of positive sessions restored. That shape is what a distribution regime looks like, and it has held since the flow peak.

Sustained weekly inflows above $500 million would mark a genuine change. August produced exactly one such week.

On-Chain Says Cheap, Not Bottomed

The valuation metrics have moved into territory that historically precedes recovery without confirming it. The MVRV Z-Score printed 0.42 on August 8, with the raw MVRV ratio at 1.24 and the aggregate realized price at $52,330. A separate reading places MVRV at 1.25 with realized price at $52,468 and the Z-Score at 0.40.

Translated: the market trades at roughly a 24% to 25% premium to its aggregate on-chain cost basis. That is a compressed premium by any cycle standard. The Z-Score has gone below zero at every previous cycle bottom, and it has not done so here. It moved up 0.10 points over the trailing 30-day window, which means the metric is drifting away from the capitulation threshold rather than toward it.

The cohort split prevents a clean read. Short-term holder MVRV sits at 0.96, below cost basis — the average buyer inside the last 155 days is underwater by roughly 4%. Long-term holder MVRV sits at 1.32, comfortably above cost basis. Net unrealized profit/loss reads 0.19, positive in aggregate but far from the euphoric readings that mark tops.

Eight of twelve capitulation signals are currently firing on the VanEck dashboard, which points to a late-stage drawdown. The caveat attached to that reading is specific: forward returns from these signal clusters have beaten the Bitcoin baseline only at the one-year horizon, on a small and heavily overlapping sample. It is a statement about where in the cycle price sits, not a timing tool.

The drawdown arithmetic frames the remaining risk. Prior bear markets bottomed 77% to 84% below the high. This one has bottomed, so far, at 54% below the October 6 peak. On duration, prior peak-to-trough spans cluster tightly at 406, 363 and 376 days. At roughly 9.7 months from the peak, this drawdown is about three-quarters through the typical duration but only about two-thirds through the typical depth.

If depth is a function of leverage and sentiment rather than market structure, the capitulation phase has not happened yet.

Long-Term Holders Dumped 356,000 Coins in Thirty Days

The most consequential on-chain development of the month is that the cohort supposed to absorb supply became the source of it. Coins held longer than one year fell by 356,000 BTC over the trailing 30 days, a 2.9% decline to 11.84 million BTC. That pushed the long-term supply share back under 60% for the first time in months.

At $64,400, 356,000 coins carries a notional value of roughly $22.9 billion. Not all of it hit the spot market — some represents custody rotation and internal transfers — but the direction is unambiguous. Holders who sat through the entire drawdown from above $126,000 chose the $63,000 to $65,000 zone to reduce.

That is what breaks range compressions. Short-term holders are already underwater at 0.96 MVRV and have limited capacity to supply more. ETF holders sit 28.6% below their $90,200 average and have been redeeming in bursts. If long-term holders are now distributing into the same band, three of the four major supply cohorts are net sellers simultaneously, and the fourth — corporate treasuries — has stopped buying entirely.

The long-term holder realized price sits near $49,700 based on mid-year data, meaning this cohort remains substantially in profit even at current levels. Their 1.32 MVRV confirms it. Profitable holders selling into a compressed range is a materially different signal than distressed holders capitulating, and it argues the floor is lower than the capitulation-signal count implies.

The short-term holder realized price around $69,007 as of mid-year is the other side of that structure. Anyone who bought the last four months of dips remains in the red, which puts a persistent overhang on every rally into the high $60,000s even if the range breaks upward.

The one cohort providing support is the aggregate realized price at $52,330. Price falling to that level would put the entire market at break-even, and every prior instance of the spot price approaching aggregate cost basis has marked a durable floor. That is 18.7% below current levels — the real downside if the range resolves badly.

Realized Volatility at 27.2% Is the Coiled Spring

Thirty-day realized volatility has collapsed to 27.2% annualized. The long-run average sits near 80%. Bitcoin is currently moving at roughly one-third of its historical pace, and it has been doing so long enough that positioning has adjusted to the calm.

Volatility compression of this magnitude does not persist. It resolves. The mechanics are straightforward: option sellers harvest premium into the quiet, dealers accumulate short gamma, and when a catalyst finally forces price outside the band the hedging flow amplifies the initial move rather than dampening it. The tighter the coil, the more violent the release.

The complicating factor is that Bitcoin options remain expensive despite the summer calm. Implied volatility has not compressed to match realized, which means the market is paying up for protection it has not needed for weeks. That gap — expensive optionality against cheap realized movement — is the market pricing a specific fear rather than general uncertainty. The fear is the long end.

The futures market structure adds to the risk. The description that keeps surfacing is a crowded club with a tiny exit: large positioning concentrated in a market whose spot liquidity has thinned considerably. Twenty-four-hour volume on the primary tape running $7.68 billion against a $1.33 trillion market cap means the order book cannot absorb a coordinated unwind without significant slippage.

Historical precedent for what negative funding plus rising open interest produces is well documented. When perpetual funding went to -6% in late February as price briefly hit $63,000, the setup resolved with a short squeeze. When the 30-day average funding rate stayed negative for 46 consecutive days through mid-April — the longest streak since the post-FTX wreckage of November 2022 — open interest rose alongside it, and that combination has historically resolved violently to the upside.

The current regime is quieter than either of those. That is precisely the problem: there is no crowded side to squeeze, which removes the mechanical fuel for a sharp reversal and leaves the range to be broken by external flow rather than internal positioning.

Strategy Stopped Buying in June and the Stock Is Down 75%

The largest corporate holder has been a source of supply, not demand, for two months. Strategy holds 840,447 BTC worth roughly $53.4 billion against a total purchase cost near $63.4 billion — an unrealized loss of approximately $10 billion, with $8.2 billion reported as a paper loss in the most recent disclosure. The implied average acquisition price is about $75,400 per coin.

The company has not purchased Bitcoin since the middle of June. Between August 10 and August 16 it sold 3.46 million MSTR shares for about $333.7 million and executed zero Bitcoin purchases or sales, keeping holdings unchanged. Of the proceeds, $52.4 million funded STRC preferred dividends, $132.2 million went to STRC repurchases under the Digital Credit Securities Repurchase Program, and $149.1 million was added to the USD reserve, lifting it to $4.8 billion. The prior week it sold 1,690 BTC outright and raised $653 million from share sales.

The framework behind this matters more than any single week. Under the Digital Credit Capital Framework adopted in late July, the USD reserve is restricted to covering preferred dividends and interest, a $1 billion repurchase program for digital credit securities was authorized with STRC prioritized, and the BTC Monetization Program was expanded to permit up to $5 billion in Bitcoin sales to fund reserves, dividends and repurchases.

That $5 billion ceiling remains almost entirely unused. A week of leaning on equity issuance instead of the treasury dilutes shareholders rather than shrinking the coin stack, which is the less damaging of the two options for Bitcoin's supply picture. It does not remove the mechanism. The capital structure requires a constant stream of financing from somewhere, and if STRC keeps trading below its $100 par value into the fall, the pressure to use the monetization authority increases.

MSTR fell more than 5% Tuesday even as Bitcoin rose, and the stock is down 75% over the trailing twelve months. The counterweight: 12 of the company's 15 largest institutional shareholders increased positions in the second quarter, lifting combined holdings by $1.2 billion.

196 Companies, One Trade, and an MSCI Deletion Risk

The corporate treasury cohort has grown to 196 public companies adopting some form of Bitcoin acquisition model, and the concentration inside it is extreme. Beyond Strategy's 840,447 BTC, the next four are Tether-backed Twenty One at 43,514 BTC, Metaplanet at 43,000 BTC, MARA at 35,577 BTC, and the Adam Back and Cantor Fitzgerald-backed Bitcoin Standard Treasury Company at 30,021 BTC. Strategy alone holds roughly five times the combined total of the next four.

Metaplanet disclosed a paper loss of $1.5 billion on its 43,000 BTC as of end-June. Combined with Strategy's $8.2 billion, the two largest holders carry close to $10 billion in unrealized losses. If those losses were tokenized, the resulting position would rank as the eleventh largest digital asset by market value.

The company spent last week defending itself against a sale that never happened. A 5,014 BTC transfer worth roughly $320 million triggered speculation, and CEO Simon Gerovich confirmed it was a routine custody operation between Metaplanet-controlled addresses with holdings unchanged at 43,000 BTC. Total network cost of moving that sum: approximately $8. On Tuesday the company agreed to transfer 2,100 BTC to Super League Enterprise to launch a U.S. treasury platform, extending the model into a Nasdaq-listed vehicle.

The index risk is the newer threat. MSCI opened a consultation proposing to identify and exclude non-operating companies from its Global Investable Market Indexes, using five financial ratios rather than any crypto-specific threshold. A simulation applied to May 2026 data would have deleted Strategy, Metaplanet and uranium holding company Yellow Cake from the MSCI ACWI IMI. Sharplink, Center Laboratories and Lydia Holding would land on a public watchlist. Any adopted changes would fold into the November 2026 index review.

Forced index selling in these names would not directly hit Bitcoin, since the underlying coins stay on balance sheets. It would compress the market-cap-to-net-asset-value ratios further — already deeply contracted from summer 2025 peaks — and close off equity issuance as a funding channel. That is the path by which an index decision becomes a spot supply event.

Global M2 at a Record and Bitcoin Down Fifty Percent

The most cited structural anomaly of 2026 is the breakdown of the relationship between Bitcoin and global liquidity. Global M2 recently surpassed a record $100 trillion, with some readings placing it near $121 trillion and rising. Over the same window Bitcoin has fallen roughly 50% from its late-2025 high above $126,000.

For years the two moved in near lockstep. The correlation was the foundation of nearly every long-horizon model in circulation. It has now been broken for the better part of twelve months, and the explanation that has gained the most traction points at AI infrastructure. Capital is rotating into data-center capital expenditure at a pace that absorbs marginal liquidity before it reaches risk assets further out the curve. The estimate that AI-related companies could issue as much as $1.5 trillion in debt this year is the same figure driving the term-premium blowout in Treasuries.

That is the mechanism connecting the 30-year at 5.338% to Bitcoin at $64,400. Both are downstream of the same capital reallocation. Money that would have found its way into a $1.33 trillion non-yielding asset is instead funding GPU clusters and the bond issuance required to pay for them. Cameron Winklevoss framed the resulting divergence as a rare entry window created by that rotation.

The counterargument from the corporate holders is arithmetic rather than tactical. Gerovich's position is that supply schedules do not change regardless of price behavior — 21 million remains 21 million while the money supply expands without limit. Fidelity's global macro director has argued that a gold and liquidity regression against global M2 implies gold worth around $5,000, and that momentum in gold would eventually carry Bitcoin and Ethereum with it.

Gold is currently doing exactly that. It ripped 1.70% to $4,495.60 Wednesday and is up more than 10% on the month while Bitcoin chops in a 3.7% band. The two assets that are supposed to share a monetary debasement thesis have decoupled completely. Until that gap closes in one direction or the other, the liquidity argument stays theoretical.

The 2:00 PM Minutes and the September Print

The July 28–29 FOMC minutes land at 2:00 p.m. ET, and they are the only scheduled event capable of resolving this range today. The committee held the federal funds rate at 3.50%–3.75% on a 9–3 vote, with three regional presidents dissenting in favor of a 25 basis point hike — the first time since September 2016 that three policymakers aligned on a single directional dissent.

Rate futures currently price a 67.4% chance of no change and a 32.6% chance of a hike at the September 15–16 meeting. Those odds have moved from a near coin flip toward a hold in recent weeks, driven by a weak July jobs report, retail sales that fell by the most in more than a year, and in-line CPI and PPI prints. CPI still runs 3.4% year over year against a 2% target.

The specific risk is that the minutes reveal hawkish sentiment extending well beyond the three named dissenters. Chair Warsh has withdrawn forward guidance entirely, which makes every meeting live and turns the minutes into the primary window on committee thinking. He has also appeared comfortable with the recent tightening in financial conditions, suggesting the market has been doing part of the Fed's work — language that, if repeated in the minutes, reads as tolerance for higher long-end yields.

A hawkish read pushes September hike odds above 40%, sends the 30-year back through 5.338%, and puts the $62,662 floor under immediate pressure. A read that isolates the dissent leaves the base case intact and gives Bitcoin room to attack $65,000 with the 10-year already down 0.21% on the session.

Historically, minutes releases trigger swift moves in rate-sensitive assets including major crypto pairs, particularly when tone diverges from the post-meeting statement. Bitcoin's correlation to the Nasdaq 100 and to real rates has been the dominant input all year.

The calendar after this is dense: July PCE on August 26, Jackson Hole from August 27 to 29 with Warsh delivering his first keynote as chair, then the September FOMC carrying a fresh Summary of Economic Projections — the first dot plot since June.

What the Betting Markets Actually Price

Prediction market odds have collapsed in a way that spot price alone does not convey. Kalshi assigns roughly a 1.6% chance of Bitcoin reaching $100,000 at any point in 2026, down from 91% in January. The probability of touching $90,000 fell from 71% in early May to 2.5%. Those are not adjustments. That is the complete removal of the upside scenario from the pricing.

The downside tail carries materially higher odds. Contracts imply a 20% chance of a decline below $45,000 and a 15% chance of a break below $40,000. On the Robinhood venue, a move under $55,000 at some point in 2026 prices at 57% — the single highest-probability directional outcome on the board.

Near-term contracts are tighter and more instructive. Bitcoin above $67,500 at any point in August prices at 26%. Below $60,000 in August prices at 19%. Bitcoin above $62,000 at 5:00 p.m. ET today prices at 99% with roughly 453,000 contracts of interest, and above $50,500 on August 21 prices at 99% across 2.6 million contracts.

Read together, the market prices this range holding through month-end with a modest skew toward the downside break coming first, and it prices essentially zero probability of a return to six figures this year. Reaching $200,000 within the next year prices at 2%. Above $100,000 in 2026 prices at 14% on one venue against Kalshi's 1.6%, a spread that reflects differing contract terms rather than genuine disagreement.

The analyst distribution mirrors that skew. One widely followed forecast calls for a decline to $20,000 by the end of 2027 before a sharp recovery. On the other side, the argument that Bitcoin's refusal to react to bad news — Strategy's sales, CLARITY Act delays, a multi-million dollar Coldcard hack, the Maya Protocol exploit that dropped pool value by $11 million, patched severe BitBox wallet flaws — constitutes a bottoming signal has gained adherents.

Santiment flagged the "crypto is dead" narrative spreading rapidly among market participants. Bitwise cut 14% of its workforce to approximately 155 people. Miners are bleeding: Cipher Digital posted a $267.5 million quarterly net loss, and BitFuFu swung to a $20.5 million loss on revenue of $42.8 million.

The Forecast: Levels, Targets and What Breaks the Range

The base case is that $62,662 holds and Bitcoin grinds toward the moving average cluster before the minutes force a resolution. The August target sits at $65,500 with a bullish extension near $67,000, and the projected 2026 range spans $59,500 to $72,097.

The bull sequence requires four consecutive reclaims: $64,147, then $64,467, then $64,557, then $65,000. All four sit within 1.3% of each other, which means a single 3% session executes the entire ladder. Above $65,000, the target is the 100-day EMA at $66,735. That level is the first genuine trend signal since the breakdown, and clearing it would be the first evidence that the marginal buyer has returned — a condition that has not been met since April 2026. Beyond $66,735 the path runs $67,500, then the $70,000 zone, then the 200-day EMA at $71,925 as the ceiling on any 2026 recovery.

The bear sequence is shorter and faster. Losing $63,174 opens $62,783 immediately. Losing $62,662 removes the floor that has held since early July and opens a 4.2% air pocket to $60,000. Below $60,000, the next structural support is the aggregate realized price at $52,330, where the entire market sits at break-even — 18.7% below spot. Prediction markets assign a 57% probability to a sub-$55,000 print at some point this year, which places that scenario inside the base case rather than the tail.

The specific triggers are known. Hawkish minutes at 2:00 p.m. that show the 9–3 vote understated committee support for tightening send the 30-year back above 5.338% and break $62,662. A read that isolates the three dissenters, combined with the 10-year already 0.21% lower, gives the $65,000 test room to work. ETF flow is the confirming variable in either direction: sustained weekly creations above $500 million would mark genuine demand returning, while a resumption of the burst-redemption pattern that produced $6.4 billion of 30-day outflows in May and June overwhelms anything the chart says.

The verdict: this is a compressed range in a corrective trend where three of four supply cohorts are net sellers, realized volatility at 27.2% guarantees a violent resolution, and the resolution is being priced by the Treasury curve rather than by anything native to Bitcoin. Base case targets $66,735 on a minutes-driven relief move. Failure at $62,662 targets $60,000 first and $52,330 as the structural floor. The 200-day EMA at $71,925 stays out of reach until the long end stops making multi-decade highs.

That's TradingNEWS