Bitcoin Daily Brief — September 8, 2026: The Golden Cross Prints at $78,447 as Band Width Falls a Tenth Day to 8.4%

Bitcoin price on September 8, 2026: BTC closed at $78,447, down 0.82%, in a $1,888 range between a $79,477 high and a $77,589 low — a session in which the 50-day moving average finally crossed above the 200-day, the first golden cross since May 2025, while price itself finished $18 from its own 20-day average and Bollinger band width contracted for a tenth straight day to 8.4%. The snapshot and charts are first, the short version follows, and the full explanations are at the end for anyone who wants them.

New to the indicators? Every name below links to the plain-English explanations in Bitcoin Indicators Explained. Yesterday’s edition: September 7, 2026. About this series: Bitcoin Daily Brief.

Bitcoin price on September 8, 2026, at a glance

The charts

The 30-second read

Bitcoin fell 0.82% on Tuesday to close at $78,447, its lowest close since September 2, in a $1,888 range on volume of 5,149 BTC — 0.67 times the twenty-day average — as US equities, gold and Bitcoin all fell together while WTI crude rose 1.9% on Middle East supply fears and CME FedWatch odds of a Federal Reserve rate increase next week were quoted between 56% and above 60%. Underneath the price the event this brief has projected for three days finally printed: the 50-day moving average closed $88.65 above the 200-day, the first golden cross since May 22, 2025 and the end of a 296-day death-cross spell, while Bollinger band width contracted a tenth straight day to 8.44% — a fifth of where it stood ten sessions ago — and the close finished $17.76 from its own 20-day average, the tightest that gap has been in percentage terms in any session of 2026. The calendar now does the talking: the delayed US spot ETF flow print for Tuesday, August producer prices on Thursday and consumer prices on Friday, and an FOMC meeting on September 15–16 whose direction is genuinely undecided.

What happened in the last 24 hours

• Price. Opened $79,092, reached $79,477, fell to $77,589 and closed $78,447 (−0.82%, −$645) — an $1,888 range on volume of 5,149 BTC, 0.67 times the twenty-day average; the low was the lowest since September 3 and the close the lowest since September 2, but both held above September 2’s $76,219 low.
• Flows. A fourth consecutive edition with no new US spot ETF figure: Farside Investors’ table still showed no published total for Tuesday September 8 when this brief was written, so Friday’s +$174.6 million stands as the last print and September’s four reported sessions stay at +$770 million against August’s +$3,524 million.
• Positioning. The only figures available were stamped at 00:06 UTC on September 8 and therefore describe the twenty-four hours into the start of Tuesday rather than Tuesday itself: $5.75 million of crypto liquidations, $4.78 million long against $976,000 short, with funding at +0.0021% per eight hours and open interest near $53.0 billion. No liquidation total for Tuesday’s own session had been reported when this brief was written.
• On-chain. Blockstream’s Liquid sidechain recovered 3,400 of the roughly 4,000 BTC withdrawn in Monday’s exploit — 85% — leaving about 600 BTC and roughly $47 million outstanding while, Decrypt reports, some observers dispute the actors’ “white hat” framing; separately Cointelegraph reported that short-term-holder whales are sitting on record unrealized gains of about $9 billion on data going back to 2016.
• Macro and policy. The S&P 500 closed at 7,690.84 (−0.36%), the Nasdaq Composite at 26,410.44 (−0.36%) and the Dow at 52,908.91 (−0.95%), with gold down 0.69% to $4,445.60 and WTI crude up 1.88% to $93.20 as Cointelegraph reported an intraday three-month high near $95; Canada’s retaliatory tariffs took effect, and CME FedWatch odds of a September 16 rate increase were quoted at about 56% by Yahoo Finance and above 60% by Motley Fool.

Section summaries

Price and trend

The golden cross printed. The 50-day average rose $264.68 to $69,956.47 while the 200-day rose $52.31 to $69,867.82, turning a $123.72 deficit into an $88.65 surplus and completing a convergence that has run for fifty consecutive sessions. It is the first golden cross since May 22, 2025 and it ends a death-cross spell that began on November 16, 2025 and lasted 296 days, the fourth-longest of the thirteen such spells in this series. Price itself went the other way, closing down 0.82% at $78,447.11 — a twenty-first straight close above the 200-day average, but the lowest close since September 2. Details ↓

Momentum and volatility

The MACD histogram widened a fifth day to −396.55 from −261.93 as the MACD line fell $234 to 2,843.71 against a signal line that eased $99 to 3,240.25 — still 2,844 above zero, so a continued loss of upward momentum rather than a change of trend direction. RSI fell 2.55 points to 60.11, its lowest since August 18, while ADX fell for the first time in twenty-one sessions, to 47.22, ending a twenty-day run of increases as the spread between the directional lines narrowed from 23.42 to 18.63. Bollinger band width contracted a tenth straight day, by 4.84 points to 8.44%, driven again by a single old close leaving the twenty-day window — and one more such close leaves on Wednesday. Details ↓

Flows and positioning

Tuesday’s ETF figure — the one carrying three days of accumulated brokerage-channel orders — had still not been published when this brief was written, so Friday’s +$174.6 million remains the most recent print and September’s four reported sessions stand at +$770 million against August’s completed +$3,524 million. The wider context is unusually strong: Cryptonomist and Crypto Briefing both report roughly $3.8 billion of net inflows over the three weeks to September 8, described as the strongest three-week run of 2026, with one week approaching $987 million. Derivatives readings were thin and stale, describing the window into the start of Tuesday rather than Tuesday’s own session. Details ↓

On-chain and network

Monday’s Liquid exploit resolved further: 3,400 of about 4,000 BTC have been returned, 85% of the total, leaving roughly 600 BTC and about $47 million outstanding while Blockstream negotiates and some observers question the “white hat” description. Cointelegraph reported that whale-sized short-term holders are carrying record unrealized gains of about $9 billion, which its analysis frames as sell-side risk rather than as strength. On the network itself difficulty is unchanged at 127.45T with the chain at block 966,124, and CoinWarz now estimates the next retarget at +2.55% to about 130.7T on September 19. Details ↓

Sentiment

alternative.me’s dated series printed 69 (“Greed”) for September 8, down two points from 71 on September 7, against 69 a week ago and 31 (“Fear”) a month ago. The news feed carried 23 Bitcoin-tagged articles, the same count as Monday, scored none bullish, five somewhat bullish, fourteen neutral, three somewhat bearish and one bearish — the single bearish score attaching to a homicide story rather than to anything about price. Published views are split between the golden-cross projections circulating at $99,000–$100,000 and CoinDesk’s own reminder that the same signal has produced premature bull traps three times as often as sustained rallies. Details ↓

Macro and cross-asset

Tuesday was a risk-off session with an energy shock inside it: the S&P 500 fell 0.36% to 7,690.84, the Nasdaq 0.36% to 26,410.44 and the Dow 0.95% to 52,908.91, gold fell 0.69% to $4,445.60, and WTI crude rose 1.88% to $93.20 after touching a three-month high near $95 on Houthi strikes in the Strait of Hormuz. Canada’s retaliatory tariffs took effect after trade talks collapsed. The live question at the September 15–16 FOMC remains whether the Fed raises, with CME FedWatch odds of an increase quoted at about 56% and above 60% by different outlets on the same day, and Bitcoin dominance was 58.3% on tv-hub.org’s measure. Details ↓

Calendar

Wednesday September 9: the delayed ETF print for Tuesday, and the last sub-$74,000 close leaves the twenty-day Bollinger window. Thursday September 10: August producer prices and weekly jobless claims. Friday September 11: August consumer prices, the report Decrypt describes both Bitcoin and the S&P 500 as waiting on. September 15–16: the FOMC, decision Wednesday at 2 p.m. ET. September 18: the September options expiry. September 19: the next difficulty adjustment, estimated at +2.55%. Details ↓

Reading the combination

Tuesday delivered a textbook bullish trend signal on a day price fell, which is the clearest possible demonstration of what a moving-average crossing actually measures: the past. The golden cross is arithmetic about closes from the last ten months, not information about the next ten days, and this series’ own record says so — twelve prior crosses since 2012, higher thirty days later six times out of twelve. Meanwhile the compression reached an extreme: band width 8.44%, %B 0.50, price $18 from its 20-day average, and a fourth straight fall in ATR. The bracketing levels are unusually tight — $79,477 up to $80,000 immediately above and $78,300 down to $77,589 below, the whole span from Tuesday’s low to the round number running about 1.06 times a single day’s ATR — and the indicators do not say which way it resolves. Details ↓

Learn the indicators: every reading above is explained, with charts and the combinations traders watch, in Bitcoin Indicators Explained.

The details, for readers who want more

Signal board

Indicator	Reading (Sep 8 close)	Zone	Change since Sep 7	What this zone has historically meant
50-day vs 200-day SMA	$69,956 vs $69,868 (gap +$88.65)	Golden cross — 50-day above, first day	Gap −$124 → +$89	The crossing completed on the session yesterday’s projection named. The 50-day rose $264.68 while the 200-day rose $52.31, a relative move of $212.37 and a fiftieth consecutive day of the shorter average gaining on the longer. It ends a death-cross spell that began on November 16, 2025 and ran 296 days — the fourth-longest of the thirteen spells in this series, behind 389, 388 and 312 days. The initial gap of $88.65 is the narrowest first-day margin of the last six golden crosses, ahead of $89.93 in May 2025 and $98.39 in September 2021, which matters only because a narrow margin can be un-crossed by a few ordinary sessions. On the record: of the twelve golden crosses in this series since 2012, price was higher ten sessions later seven times (median +1.4%), twenty sessions later seven times (median +4.1%), thirty sessions later six times (median −2.6%), sixty sessions later eight times (median +12.6%) and ninety sessions later eight times (median +10.7%). That is a signal with a positive tilt at two and three months and no edge at all at one, on a sample of twelve.
Price vs 200-day SMA	$78,447 vs $69,868 (+12.3%)	Above (21st straight close, since Aug 19)	+13.3% → +12.3%	Price above the 200-day average is the most common definition of a long-term uptrend, and this is a twenty-first consecutive close above it. The distance narrowed by a further point, again with price doing most of the work: the 200-day average rose $52 while price fell $645. The 200-day has now been rising for long enough that the crossing above it by the 50-day became arithmetically likely; the two facts are the same fact seen at different horizons.
Price vs 20-day SMA	+0.02% above ($78,429)	Effectively on the average	+1.4% → +0.02%	The close finished $17.76 above the 20-day average — the smallest gap between close and 20-day average, measured as a percentage of price, of any session in 2026. Price fell $645 while the average rose $457, and they met. A short average sitting exactly under the close is the definition of a market with no short-term direction; it is also the state from which the 20-day line stops acting as support or resistance and simply becomes the price.
RSI (14)	60.11	Upper neutral zone (30–70)	62.66 → 60.11	Down 2.55 points and the lowest reading since August 18, when it was 55.58 on the eve of the breakout. RSI compares the average size of up-days with down-days over fourteen sessions; a $645 fall is a small input by the standards of the last three weeks, which is why the decline was modest. Of 351 earlier days in this series with RSI between 58 and 62 and price above its 200-day average, the close was higher ten sessions later 56% of the time (median +1.3%) and twenty sessions later 54% (median +2.5%). Readings in the sixties are the ordinary resting state of an uptrend, not a signal in either direction.
MACD (12, 26, 9)	2,844 vs signal 3,240; histogram −396.55	Below signal, far above zero	Histogram −261.93 → −396.55 (widening)	A fifth day below the signal line and a fifth consecutive widening, the sequence now reading −49.50, −108.96, −139.48, −261.93, −396.55. The MACD line fell $234 while the signal line eased $99. The line is still 2,844 above zero, so this describes a loss of upward momentum rather than a change of trend direction — and it sits oddly beside the golden cross, because the two indicators are reading different windows of the same data. Of 636 earlier days in this series with a negative and widening histogram, the MACD line above zero and price above its 200-day average, the close was higher ten sessions later 55% of the time (median +0.8%) and twenty sessions later 53% (median +0.8%).
Bollinger Bands (20, 2)	%B 0.50; band width 8.44%	Mid-band; squeeze deepening	%B 0.61 → 0.50; width 13.3% → 8.4%	A tenth straight contraction, of 4.84 points, in a run that now reads 0.70, 1.03, 1.72, 2.16, 2.16, 3.30, 4.43, 4.52, 6.74 and 4.84. Band width is now 21% of the 40.03% peak recorded on August 29, ten sessions ago. The mechanism is the same one as Monday’s and it is worth repeating: August 19’s close of $69,300 left the twenty-day window, taking the twenty-day standard deviation from $2,588 to $1,655, so the upper band fell $1,409 to $81,739 while the lower band rose $2,323 to $75,120. %B at 0.5027 puts the close almost exactly at the middle band. Two base rates apply and they disagree: of 268 earlier cases of band width contracting ten days in a row with price above the 200-day average, the close was higher ten sessions later 49% of the time (median −0.3%) and twenty sessions later 52% (median +1.0%); but of 510 earlier days with band width at or below 9% and price above the 200-day, it was higher ten sessions later 62% of the time (median +1.4%) and twenty sessions later 65% (median +4.2%). The contraction itself is neutral; the low absolute level has leaned mildly positive.
ADX (14)	47.22 (+DI 31.7, −DI 13.1)	Strongly trending (above 25), first fall in 21 sessions	47.65 → 47.22	The twenty-day run of increases ended. ADX fell 0.43 points as the internals continued to deteriorate: +DI fell from 33.71 to 31.70 while −DI rose from 10.29 to 13.07, narrowing the spread from 23.42 to 18.63 points, the second consecutive narrowing. ADX is a fourteen-period smoothing of that spread, so it turns well after the spread does, and the turn has now arrived. Days in this series with ADX at or above 45, +DI above −DI and ADX falling (253 instances) were followed by a higher close ten sessions later 71% of the time (median +4.7%) and twenty sessions later 70% (median +5.6%) — the strongest reading on this board, with the standing caveat that such instances cluster in a handful of strongly trending years.
ATR (14)	$2,267 (2.9% of price)	Falling, fourth straight day	$2,296 → $2,267	A fourth consecutive fall, of $29, on an $1,888 high-to-low range that was 0.822 times Monday’s ATR. That range ranks 99th narrowest of 2026’s 251 sessions by dollars and 82nd by percentage of the closing price, so it was a slightly-below-average day rather than a quiet one — and it was 5% wider than Monday’s. The fall in ATR is the fourteen-day window continuing to shed the large days of the mid-August breakout.
30-day realized volatility	47.2% annualized	Elevated, at a five-month high	47.0% → 47.2%	Up 0.21 points, a second consecutive rise, and the point worth holding on to: while Bollinger band width has fallen by four-fifths in ten sessions, the actual dispersion of daily returns over thirty days has not fallen at all. The two measures use different windows, and the band-width collapse is dominated by which closes are inside a twenty-day frame rather than by how much the market is moving.
Volume vs 20-day average	5,149 BTC vs 7,681 average (0.67×)	Below average	3,293 → 5,149 BTC	Up 56% from Monday’s holiday session, as US markets reopening would imply, but still the 73rd lightest of 2026’s 251 sessions. The twenty-day average itself fell from 8,270 to 7,681 as the heavy late-August days rolled out of it. Of 456 earlier days in this series with volume below 0.70 times its twenty-day average and price above the 200-day average, the close was higher ten sessions later 57% of the time (median +1.1%) and twenty sessions later 56% (median +2.0%).
US spot ETF net flows	Not yet reported when this brief was written; last was Fri Sep 4 at +$174.6M; September ≈ +$770M; August +$3,524M	Fourth edition without a new figure	No published change	Unlike the previous three days, Tuesday was a normal US business day, so a figure exists — it had simply not been published on Farside Investors’ table at the time of writing, which is ordinary: the settled per-fund numbers typically appear the following day. This is therefore a reporting lag rather than the absence of a session, which is the opposite of the situation over the holiday weekend. When it arrives it will be the first read on whether three days of accumulated brokerage-channel orders arrived as demand or as supply, and it lands two days before the first of the week’s two inflation reports. The wider flow picture is strong on any reading: Crypto Briefing and Cryptonomist both report roughly $3.8 billion of net inflows over the three weeks to September 8, the strongest three-week stretch of 2026, with one week approaching $987 million.
Perpetual funding rate	+0.0021% per 8h (≈ +2.30% annualized)	Positive, well below the +0.01% “crowded long” threshold	+0.0028% → +0.0021% on the same source, but see the timing note	Funding is the periodic payment between holders of perpetual futures that keeps the contract tethered to spot; positive means longs pay shorts. The same timing problem as yesterday applies, one day later: the CoinStats page this brief reads carried an “08 Sep, 2026 00:06” stamp, so its readings describe the twenty-four hours into the start of Tuesday rather than Tuesday’s own session, and the prices it quotes, $79,115 to $80,055, straddle Monday’s close rather than Tuesday’s $78,447. It is reported for continuity with that caveat attached. At +0.0021% per eight hours it annualizes to +2.30% on the ×3×365 convention used throughout this brief — positive but modest, and a step down from Monday’s reading on the same source.
Open interest / liquidations	24-hour liquidations $5.75M across crypto; OI ≈ $53.0B on CoinStats	83% long; stale window	Liquidations $208M → $5.75M on different sources and windows	Liquidations are forced closures of leveraged positions by an exchange when margin runs out. The only figure available describes the window ending at the start of Tuesday: $5.75 million across crypto, $4.78 million long (83%) against $976,000 short. That is a very small number — it says leverage was barely tested in the hours after Monday’s close — and it cannot be compared with Monday’s $208 million, which came from a different tracker over a different window. No total for Tuesday’s own session had been published when this brief was written, which matters because Tuesday’s $1,888 range and $645 fall are exactly the conditions that produce long liquidations. On open interest the level is reported but no day-over-day change is claimed, for a sixth consecutive day: the same tracker prints between $52.96 billion and $53.01 billion depending on where on the page it is read.
Fear & Greed	alternative.me 69 (Greed)	Greed	71 → 69	Down two points on the index’s own dated series, read from its API rather than from the front page, and a second consecutive two-point fall. The dated values run 74 (Sep 4), 73 (Sep 5), 73 (Sep 6), 71 (Sep 7), 69 (Sep 8), with 69 a week ago on September 1 and 31 (“Fear”) a month ago on August 9. The value dated a given day is struck at the start of that UTC day, half a day behind the price candle used throughout this brief, which closes at the end of the same UTC day — so the September 8 reading of 69 is largely a verdict on Monday’s session. The index blends volatility, momentum, dominance and survey inputs rather than tracking price; note that a collapse in volatility of the kind chart 3 shows pushes it up mechanically, so a two-point fall against that tailwind is a slightly weaker reading than it looks. CoinStats’ own page shows 70 for the same period, which is a different index construction rather than a contradiction.
Hashrate / difficulty	Difficulty 127.45T; block height 966,124	Between retargets; estimate firmed slightly	Next-adjustment estimate +2.63% → +2.55%	Difficulty is unchanged and will stay so until the next 2,016-block boundary at height 967,680, now 1,556 blocks away with the cycle 460 blocks in. CoinWarz trimmed its projection again, from +2.63% to +2.55%, to about 130.7T, and now times it at 12:52 p.m. UTC on September 19 — roughly two hours later than yesterday’s estimate. A positive estimate means blocks are arriving faster than the ten-minute target, which is what a rising hashrate produces. The site did not display a headline hashrate figure when it was read for this edition, so no reading is quoted; yesterday’s brief flagged its 23% one-day swing as a probable estimation artefact, and the September 19 retarget will settle the question in a way no estimate can, because difficulty adjusts on what actually happened over 2,016 blocks.

Price and trend (chart 1)

The event arrived, and price did not celebrate it. Bitcoin opened at $79,091.97 — Monday’s close, as the daily candles here are continuous — reached $79,477.07 in the early hours, fell through the US session to $77,589.00 and closed at $78,447.11, a loss of $644.86 or 0.815%. The range was $1,888.07, or 2.41% of the closing price and 0.822 times Monday’s ATR; that ranks 99th narrowest of 2026’s 251 sessions by dollar range and 82nd by percentage of the close, so it was a slightly-below-average day, and it was 5% wider than Monday’s $1,796. The candle’s shape is worth reading: the body was $644.86, only 34.2% of the range, with an $858.11 lower wick and a $385.10 upper wick (a wick is the thin line above or below the candle body, marking where price traded but did not close), and the close finished 45.4% of the way up the range. A body that small relative to its wicks describes a session that probed in both directions and settled near the middle — the visual signature of indecision rather than of distribution. Volume was 5,149 BTC against a twenty-day average of 7,681, a ratio of 0.670, up 56% from Monday’s holiday reading as US markets reopened but still the 73rd lightest session of the year.

Two boundaries moved and one held. The low of $77,589.00 undercut both Monday’s $78,666 and Friday’s $78,626, making it the lowest low since September 3, and the close of $78,447.11 is the lowest close since September 2’s $77,307.36 — so the observation this brief made on Monday, that the September range had not been extended downward, no longer holds at the four-session horizon. What does still hold is the month’s floor: September 2’s low of $76,219.18 is $1,370 below Tuesday’s low and has not been approached. Against longer reference points the close sits 37.9% below the all-time high of $126,296, 10.3% below the 2025 year-end close of $87,498, and 34.0% above the June 30 low close of $58,523.93. September is 0.15% lower across its first eight sessions, having been 0.7% higher after seven — a month that has finished negative in nine of the sixteen completed Septembers in this record and is the weakest calendar month in the series by average return, at −3.3%, though the last three Septembers all closed positive at +4.0%, +7.3% and +5.4%. On the weekly frame, the 50-week simple moving average computed here stands at $80,343.38, and Tuesday’s close is $1,896 below it; the next weekly close is struck at 00:00 UTC on Monday September 14.

The convergence completed. The 50-day average rose $264.68 to $69,956.47 while the 200-day rose $52.31 to $69,867.82, turning Monday’s $123.72 deficit into an $88.65 surplus — a relative move of $212.37 and a fiftieth consecutive session of the shorter average gaining on the longer. Yesterday’s mechanical projection, which held price flat and rolled both windows forward, put the crossing on this session by about $98; it landed at $88.65, and the small shortfall is entirely explained by price not staying flat. This is the thirteenth golden cross in this series since 2012 — the twelve prior ones supply the base rates below — and the first since May 22, 2025. It ends a death-cross spell that began on November 16, 2025 and ran 296 days, the fourth-longest of the thirteen spells this record contains, behind 389 days from January 2022, 388 from March 2018 and 312 from September 2014. Bitcoin closed 12.3% above the 200-day average, 12.1% above the 50-day, 17.8% above the 100-day and 0.02% above the 20-day — a twenty-first consecutive close above the 200-day in the run that began on August 19.

What the crossing is worth is a separate question from whether it happened, and the honest answer is: less than the coverage suggests, and not nothing. Of the twelve golden crosses in this series since 2012, price was higher ten sessions later seven times (median +1.4%), twenty sessions later seven times (median +4.1%), thirty sessions later six times out of twelve (median −2.6%), sixty sessions later eight times (median +12.6%) and ninety sessions later eight times (median +10.7%). The shape of that is a coin flip at one month and a mild positive tilt at two and three, on a sample small enough that a single outcome moves it. CoinDesk, running the same exercise on its own data, counts twelve occurrences in Bitcoin’s history, finds an average three-month gain of 24.9% across nine measurable instances, and notes that only three of the twelve remained valid a full year — its summary being that the signal “triggered premature bull traps 3x more often than sustained multi-year rallies” and that this one is “far from a guarantee of long-term bullishness.” The count difference between that tally and this one is a data-start artefact of a different kind: this series begins on July 17, 2010, and on the first date for which both averages can be computed — February 1, 2011 — the 50-day is already above the 200-day, so the record opens mid-uptrend rather than at a crossing. Two structural points are worth adding. First, a golden cross is entirely backward-looking: the 50-day average on September 8 is the mean of closes from July 21 onward, and the 200-day the mean of closes since late February, so the crossing is a statement about the summer, not about the autumn. Second, the initial margin here is thin — $88.65, the narrowest first-day gap of the last six golden crosses — and a margin that thin can be reversed by a handful of ordinary down-days, which is why the crossing is better read as a threshold quietly passed than as an event with a date.

Levels being discussed (attributed)

Level	Why it is being watched	Who cites it
$300,000 by 2029	A long-horizon house forecast, quoted here only to mark the range of published views	Bernstein, via Motley Fool (Sep 8)
$99,000–$100,000	A projection formed by applying the historical 24.9% average three-month gain after a golden cross to the current price; the same coverage calls it “plausible but far from certain”	Crypto Briefing and Cryptonomist (Sep 8)
$83,000–$86,000	Dense cluster of short liquidation levels above the market — the band where forced buying would concentrate if price rose into it	CryptoQuant, via KuCoin daily report (Sep 7)
$82,283	The September 3 high, the highest since May 11, and the level analysts describe as having rejected price four times since August 25	Alpha Vantage feed; 24/7 Wall St (Sep 7)
$82,240	“Main resistance obstacle” on the four-hour chart	CobraTrader, via CoinTurk (Sep 8)
$81,809	The 365-day moving average, computed here	This brief’s calculation
$81,739	Upper Bollinger Band, 4.2% above the close and $1,409 lower than yesterday	This brief’s calculation
$81,041	The 50-week moving average on Galaxy Digital’s measure, described as separating longer uptrends from downtrends; this brief’s own 50-week SMA computes to $80,343	Galaxy Digital, via 24/7 Wall St (Sep 7); this brief’s calculation
$80,343	50-week simple moving average, computed here; Tuesday’s close is $1,896 below it	This brief’s calculation
$80,000	Round number, now given back on a closing basis for a second consecutive session	Widely cited; RioTimes market wrap (Sep 8)
$79,477	Tuesday’s high	Alpha Vantage feed
$78,429	20-day simple moving average, $18 under the close	This brief’s calculation
$78,300	Support zone described as under active retest; a weekly close below it “followed by a bearish retest just like in early May would likely confirm a breakdown”	Rekt Capital, via Cointelegraph (Sep 8)
$77,589	Tuesday’s low, the lowest since September 3	Alpha Vantage feed
$77,307	The September 2 close, named as potential support if the CLARITY Act cloture vote fails and the Fed raises	24/7 Wall St (Sep 7)
$76,219	The September 2 low — the floor of the month’s range, $1,370 below Tuesday’s low	Alpha Vantage feed
$76,053 / $75,595	Retracement levels inside the short-term support zone	CobraTrader, via CoinTurk (Sep 8)
$75,500–$76,000	Short-term support zone; the same analysis sketches consolidation near $78,000–$79,000 before a possible dip into it	CobraTrader, via CoinTurk (Sep 8)
$75,120	Lower Bollinger Band, $2,323 higher than yesterday	This brief’s calculation
$72,762	200-day exponential moving average	This brief’s calculation
$72,613	Trend support named for a deeper-correction scenario	Jesse Olson, developer of Markets Sniper, via CoinTurk (Sep 8)
$69,956 / $69,868	50-day and 200-day simple moving averages — the two lines that crossed on Tuesday, now $89 apart	This brief’s calculation
$65,185	Long-term moving-average support described as preserving the bullish structure	Jesse Olson, via CoinTurk (Sep 8)
$60,000–$63,000	The nearest dense band of long liquidation levels below the market	CryptoQuant, via KuCoin daily report (Sep 7)

One note on the table. Nine of these twenty-four entries are dated September 8, and the day’s commentary clustered unusually tightly around the price rather than around the crossing: Rekt Capital’s $78,300 sits $147 under Tuesday’s close and CobraTrader’s $78,000–$79,000 consolidation sketch brackets it on both sides, while the golden-cross coverage mostly skipped to a three-month projection near $100,000 without naming an intermediate level at all. Stripping out the two liquidation-map entries and the long-horizon forecasts, the working band analysts are actually naming is $75,500 to $82,240 — CobraTrader’s floor and ceiling — which is $6,740 wide, or about 2.97 times Tuesday’s ATR. The tighter bracket that Tuesday’s own trade defined is much narrower: $77,589 to $79,477, a span of $1,888, or 0.83 times ATR.

Momentum and volatility (charts 2 and 3)

Momentum kept leaking on the day the trend signal fired, and the contrast is the most instructive thing on chart 2. The MACD line fell to 2,843.71 from 3,077.46 while the signal line eased to 3,240.25 from 3,339.39, widening the histogram to −396.55 from −261.93 — a fifth session below signal, in a sequence that now runs −49.50, −108.96, −139.48, −261.93, −396.55, widening on every one of the five. The MACD line remains 2,844 above zero, so this is a loss of upward momentum rather than a change of trend direction, and the mechanism is unchanged: the twelve-day exponential average still carries the memory of September 3’s $81,264 close and is shedding it faster than the twenty-six-day average is. That a golden cross and a five-day widening negative MACD histogram can print on the same session is not a contradiction — MACD reads twelve and twenty-six days, the moving-average crossing reads fifty and two hundred, and a market can be losing short-term momentum inside a structure that is still turning up on the long frame. Of 636 earlier days in this series with the histogram negative and widening, the MACD line above zero and price above its 200-day average, the close was higher ten sessions later 55% of the time (median +0.8%) and twenty sessions later 53% (median +0.8%). RSI fell to 60.11 from 62.66, a 2.55-point drop and its lowest since August 18. ADX did something it had not done in twenty-one sessions: it fell, to 47.22 from 47.65, ending a twenty-day run of increases. The internals had been signalling that turn for two days — +DI fell from 33.71 to 31.70 while −DI rose from 10.29 to 13.07, narrowing the spread from 23.42 to 18.63 points, its second consecutive narrowing — and because ADX is a fourteen-period smoothing of that spread, it turns last. A falling ADX from a high level describes a strong trend losing intensity, not a trend reversing.

The volatility reading did exactly what yesterday’s edition said it would, and for exactly the reason given. Bollinger band width contracted for a tenth straight day, to 8.44% from 13.28% — a fall of 4.84 points, the second-largest of the run, whose daily contractions now read 0.70, 1.03, 1.72, 2.16, 2.16, 3.30, 4.43, 4.52, 6.74 and 4.84. From the 40.03% peak of August 29, ten sessions ago, band width is down to 21% of its value. The arithmetic underneath is once again a single number leaving the window: August 19’s close of $69,300 — the last close below $70,000 in the twenty-day frame — rolled out, taking the twenty-day standard deviation from $2,588 to $1,655, so the upper band fell $1,409 to $81,738.97 while the lower band rose $2,323 to $75,119.73. This is the second consecutive day on which the squeeze deepened not because the market went quiet but because a twenty-day formula forgot the August breakout, and the proof is in the third line of the signal board: thirty-day realized volatility rose again, from 47.04% to 47.25%. The two measures are not in conflict; they are looking through different windows at the same data, and only one of them is being rearranged by the calendar. One more such rearrangement is due: August 20’s close of $73,011.87 is now the lowest in the window and it leaves after Wednesday’s session, which on a flat close would take band width to about 5.6%. The other volatility measures are milder. ATR fell a fourth day, by $29 to $2,267.29, on a range that was 0.822 times Monday’s average. %B at 0.5027, against 0.608, puts the close almost exactly at the middle band, in a band now running $75,120 to $81,739 — a $6,619 envelope against $10,351 a day ago. The base rates on the squeeze pull in opposite directions and both belong here: of 268 earlier cases of band width contracting ten days in a row with price above the 200-day average, the close was higher ten sessions later 49% of the time (median −0.3%) and twenty sessions later 52% (median +1.0%) — no edge; but of 510 earlier days with band width at or below 9% and price above the 200-day, the close was higher ten sessions later 62% of the time (median +1.4%) and twenty sessions later 65% (median +4.2%). The distinction those two rates draw is between the process of contracting, which carries no information, and the state of being tightly compressed above a rising long-term average, which historically has leaned mildly positive. Neither says anything about direction on any particular day.

Flows and positioning (chart 4)

For a fourth consecutive edition there is no new ETF number, but the reason has changed and the change matters. Over Saturday, Sunday and Labor Day Monday there was no figure because there was no primary-market session — a gap in the world, not in the data. Tuesday was an ordinary US business day, so creations and redemptions did occur; the settled per-fund figures had simply not been published on Farside Investors’ table when this brief was written, which is the normal one-day reporting lag. Friday’s print of +$174.6 million therefore remains the most recent, with BlackRock’s IBIT at +$117.4 million, about 67% of it, Fidelity’s FBTC at +$57.2 million, and every other US spot Bitcoin ETF at exactly zero. September’s four reported sessions stand at about +$770 million against August’s completed +$3,524 million, and chart 4 is unchanged for a fourth edition. The medium-term picture is the strongest part of the flow story and it is not in dispute: Crypto Briefing and Cryptonomist both report roughly $3.8 billion of net inflows across the three weeks to September 8, described as the strongest three-week stretch of 2026, with a single week approaching $987 million and institutional capital arriving “at the fastest clip of 2026.” Set against a price that is 0.15% lower on the month, that is the tension worth holding: $3.8 billion of net demand over three weeks has bought a market that has gone sideways, which either means the demand is being absorbed by supply that is not visible in ETF data, or that the flows and the price are simply operating on different clocks. Tuesday’s delayed print carries three days of accumulated brokerage-channel orders and arrives two days before the first of the week’s two inflation reports, which gives it more weight than a single day’s figure usually has.

The derivatives picture is thin and stale, and it is reported with that stated plainly rather than dressed up. The CoinStats page from which this brief takes funding, open interest and liquidations carried an “08 Sep, 2026 00:06” stamp, and the prices it quotes — a $79,115 to $80,055 spread depending on source — straddle Monday’s close rather than Tuesday’s $78,447. Its readings therefore describe the twenty-four hours into the start of Tuesday: liquidations of $5.75 million across crypto, $4.78 million long (83%) against $976,000 short; funding at +0.0021% per eight hours, about +2.30% annualized on the ×3×365 convention used here; and open interest between $52.96 billion and $53.01 billion depending on where on the page it is read. The $5.75 million liquidation figure is genuinely small — it says leverage was barely tested in the quiet hours after Monday’s close — but it is not a reading on Tuesday, whose $1,888 range and $645 decline are precisely the conditions that generate forced selling of long positions. No liquidation total for Tuesday’s own session had been published anywhere this brief could source when it was written, so none is quoted; that is a gap, and it will most likely be filled by Wednesday’s coverage. For a sixth consecutive day no day-over-day open-interest change is claimed, because the same tracker reports mutually inconsistent levels and rates of change that cannot be reconciled from the outside. On the broader market, Bitcoin’s share of total crypto value was 58.3% on tv-hub.org’s whole-market measure dated September 8, against a total crypto market value of about $2.70 trillion; the same page notes that stripping out more than $300 billion of stablecoins would raise the reading by roughly six to eight percentage points. One caution about that source: it also quotes a Bitcoin market value near $1.39 trillion, which is 51.5% of the $2.70 trillion total rather than 58.3%, and at Tuesday’s close and roughly 19.95 million coins in issue the market value computes nearer $1.57 trillion. The headline share is reported here as published, with the discrepancy flagged rather than reconciled. CoinStats’ own dominance figure of 57.6% for the overlapping window is a differently constructed measure rather than a contradiction.

On-chain and network

Monday’s Liquid incident moved toward resolution without quite reaching it. Cointelegraph reported that the actors returned 85% of the Bitcoin withdrawn from the Blockstream-operated sidechain’s federation wallet, and Decrypt put the arithmetic precisely: 3,400 of the roughly 4,000 BTC drained have come back, leaving about 600 BTC — roughly $47 million — outstanding, with Blockstream still negotiating and some observers disputing the “white hat” framing the actors have claimed for themselves. The network has restarted. As noted yesterday, a 4,000 BTC movement is not a supply event of a size that touches price, and the great majority of it came back within a day; what it does affect is coverage tone, and Tuesday’s sentiment tally shows the residue. The day’s more analytically interesting on-chain item came from Cointelegraph’s markets desk: unrealized gains held by whale-sized short-term holders — speculative wallets that acquired their coins within the last 155 days — hit their highest level on record last week at about $9 billion, on data going back to 2016. The article’s framing is sell-side risk rather than strength, and the logic is worth stating because it is not obvious: a cohort sitting on record paper profits has both the motive and the means to realize them into any strength, so a record reading is a description of overhang, not of conviction. It is a condition, not a trigger, and nothing in it says when or whether that supply arrives.

Corporate treasuries pulled in two directions on the same day, which is itself the story. Strategy — the largest corporate holder, at 845,050 BTC — skipped a Bitcoin purchase for the first time in a week and instead repurchased $176.3 million of its own STRC preferred stock, doubling the authorization on its digital securities repurchase programme to $2 billion; Decrypt’s headline that “Strategy’s return to Bitcoin buying lasted exactly one week” captures the shift, and a company choosing to buy back its own paper rather than the asset behind it is a data point about relative value as that company sees it. Pulling the other way, Strive, the asset manager founded by Vivek Ramaswamy, added 1,375 BTC for about $109 million, a third consecutive week of 5%-plus growth in its holdings, and the French listed company Capital B’s 376 BTC purchase from Monday still sits in the tape. Net of all that, corporate demand continued but its largest single source paused. On the network itself, difficulty is unchanged at 127.45T and will stay there until the next 2,016-block boundary at height 967,680, now 1,556 blocks away with the chain at 966,124 and the cycle 460 blocks in. CoinWarz trimmed its retarget estimate again, from +2.63% to +2.55% — to about 130.7T — and now times it at 12:52 p.m. UTC on Saturday September 19. A positive estimate means blocks are still arriving faster than the ten-minute target. The site did not display a headline hashrate reading when it was read for this edition, so none is quoted here; yesterday’s brief flagged its 23% one-day swing as a probable artefact of the estimation window, and that judgment stands unrevised until the retarget itself provides an answer that is measured rather than inferred.

Sentiment

The alternative.me Fear & Greed Index printed 69 (“Greed”) for September 8, down two points from 71 on September 7, taken from the index’s dated API series rather than from its front page. The dated values now run 74 on September 4, 73 on September 5, 73 on September 6, 71 on September 7 and 69 on September 8, against 69 a week ago on September 1 and 31 (“Fear”) a month ago on August 9. Two points about how to read it, repeated because they matter to the arithmetic. The value dated a given day is struck at the start of that UTC day, half a day behind the price candle used throughout this brief, which closes at the end of the same UTC day — so the September 8 reading of 69 is largely a verdict on Monday’s session, and Tuesday’s own fall will show up in the September 9 entry. And the index blends volatility, momentum, market dominance and survey components rather than tracking price. That second point cuts in a specific direction this week: a collapse in volatility of the kind chart 3 shows raises the index mechanically, so two consecutive two-point declines against that tailwind describe a slightly weaker sentiment picture than the raw numbers suggest. The index has now fallen on three of the four days since its 74 reading of September 4, holding flat on September 6, while remaining comfortably inside “Greed”. CoinStats’ own page carries 70 for the overlapping period; that is a differently constructed index rather than a contradiction, and this brief does not reconcile the two.

Of the 23 Bitcoin-tagged articles in the Alpha Vantage news feed published between 00:00 UTC on September 8 and 04:00 UTC on September 9 — the same count as Monday — the feed’s sentiment model scored none as bullish, five as somewhat bullish, fourteen as neutral, three as somewhat bearish and one as bearish. The sourcing was more balanced than Monday’s: eight from Cointelegraph, eight from Decrypt, five from Motley Fool, and one each from Zacks and Business Insider. The distribution is more neutral-weighted than Monday’s and the tail is longer at the bearish end, but the composition undercuts any market reading of it: the single bearish score attached to a Cointelegraph report on a quadruple homicide in Mexico tied to an alleged Bitcoin robbery, and two of the three somewhat-bearish scores went to a New York town weighing a mining and data-centre moratorium and to a wrap noting the day’s fall. Only one negative score in the set is about market conditions at all. The most sharply negative ticker-level score in the feed, meanwhile, belongs to Cointelegraph’s own report that Bitcoin faces a support battle at $78,300, which the model marked bearish on the Bitcoin ticker while scoring the article somewhat-bearish overall — a reasonable read, and the one piece in the set that is genuinely about price. A tally of this kind measures the tone of coverage as a machine reads it, and it is reported for continuity rather than as a market indicator.

Published views, in both directions. The constructive case ran almost entirely through the golden cross, and it ran fast: Crypto Briefing and Cryptonomist both took the historical average three-month gain of 24.9% after prior crosses and projected $99,000–$100,000, noting that the last three crossings preceded rallies of 50%, 45% and 60%, while calling the target “plausible but far from certain.” Bernstein’s long-horizon $300,000 by 2029 circulated again via Motley Fool. The cautious case came, unusually, from the same coverage: CoinDesk’s own analysis of the crossing points out that only three of twelve historical instances remained valid a full year, that the signal has “triggered premature bull traps 3x more often than sustained multi-year rallies,” and that this one is “far from a guarantee of long-term bullishness.” On the chart side, Rekt Capital identified $78,300 as under active retest and set an explicit condition — “a Weekly Close below $78300 followed by a bearish retest just like in early May would likely confirm a breakdown” — which is the most falsifiable statement any analyst made on the day. CobraTrader sketched consolidation at $78,000–$79,000 before a possible dip to $75,500, with $82,240 as the resistance obstacle. Jesse Olson described an unconfirmed buy signal on the two-week chart, similar to the pattern before the 2023 recovery, with momentum rebounding from 33.54 to about 47.40, and named $65,185 and $72,613 as the structural supports beneath — adding that the bullish case only solidifies if the signal confirms on the next candle close. Set against all of it, the macro voices were the bearish ones: the argument running through Tuesday’s wraps is that a Fed that raises rates next week lifts bond yields and redirects capital away from an asset that pays no yield, with core PCE at 3.3% and August payrolls strong enough to keep that on the table.

Macro and cross-asset

US markets reopened after Labor Day into a risk-off session with an energy shock inside it, and Bitcoin traded with the risk assets rather than against them. The S&P 500 closed at 7,690.84, down 27.76 points or 0.36%; the Nasdaq Composite at 26,410.44, down 96.55 or 0.36%; and the Dow at 52,908.91, down 505.34 or 0.95%, the widest of the three declines. Gold fell 0.69% to $4,445.60. The exception was crude: WTI rose 1.88% to $93.20, with Cointelegraph reporting an intraday three-month high near $95 and Brent quoted toward $100, a level last seen on July 24, after Houthi strikes and US–Iran exchanges around the Strait of Hormuz. Canada’s retaliatory tariffs took effect on Tuesday morning after trade negotiations collapsed. Closing Treasury yields for Tuesday could not be sourced when this brief was written and are therefore not quoted; the only fixed-income observation available was a note that the two-year yield had reached a 52-week high, carried on a page whose timestamp places it at the start of the day rather than its close. At Tuesday’s marks Bitcoin buys about 17.65 ounces of gold. No clean day-over-day comparison is available: yesterday’s edition carried gold at Friday’s $4,513.20 because the metal had no Monday US close to mark against, so the 17.52 ratio quoted then and the 17.65 quoted now are struck on bases three days apart rather than one.

The policy backdrop remains the unusual part of this cycle: the live question at the September 15–16 FOMC is whether the Fed raises rates, with the target range at 3.50%–3.75%. Tuesday’s quoted odds of an increase spanned a wide band on the same underlying market — Yahoo Finance put CME FedWatch at “nearly 56%”, Motley Fool at “over 60%”, with Kalshi at 48% and Polymarket at 49% on the same page — and that dispersion is itself informative: it describes a meeting the market genuinely cannot call. Yahoo notes that before Chair Kevin Warsh’s Jackson Hole remarks the odds of unchanged rates were near 70%, so the repricing is recent. That is why this week’s two inflation prints matter more than usual: August producer prices land Thursday and August consumer prices Friday, and Decrypt’s Tuesday wrap described both Bitcoin and the S&P 500 as “waiting on Friday’s inflation report before the Fed’s September 16 rate call” — with the S&P grinding inside its tightest range yet, the same compression story chart 3 tells for Bitcoin. Tuesday’s oil move complicates that directly: higher energy prices feed headline inflation, and The Kobeissi Letter’s note that “inflation expectations continue to mount as a result” of rising diesel prices is the link between the Strait of Hormuz and the September 16 decision. The transmission to Bitcoin is the one the coverage keeps making — higher policy rates raise the yield on dollar cash and reduce the relative appeal of an asset that pays none — and it explains why a session in which a bullish trend signal fired still closed lower.

Calendar: what’s ahead

Date (ET)	Event	Why Bitcoin traders watch it
Wed, Sep 9, morning	US spot ETF flow print for Tuesday September 8	Three days of accumulated brokerage-channel orders in one figure, delayed by the normal reporting lag
Wed, Sep 9, 8 p.m.	August 20’s $73,012 close leaves the 20-day Bollinger window	The last sub-$74,000 close in the frame; on a flat close it would take band width to roughly 5.6%, a third mechanical tightening in three days
Thu, Sep 10, 8:30 a.m.	August producer prices and weekly jobless claims	The first of the two inflation prints before the FOMC, and the first read on whether the oil move is reaching the pipeline
Thu, Sep 10, ~8:15 a.m.	European Central Bank rate decision and press conference	The euro side of the dollar index, which has tracked Bitcoin inversely through this run
Fri, Sep 11, 8:30 a.m.	August consumer price index	The report Decrypt describes both Bitcoin and the S&P 500 as waiting on; four days before the FOMC convenes
Fri, Sep 11, 10:00 a.m.	University of Michigan consumer sentiment, preliminary	Carries the inflation-expectations series that a $95 oil price would show up in first
Sun, Sep 13, 8 p.m.	Weekly candle closes (00:00 UTC Monday)	The level Rekt Capital’s breakdown condition is defined on, and the next test of the 50-week average at about $80,343
Tue, Sep 15	Senate cloture vote on the CLARITY Act	Cloture is the vote to end debate and move to a final vote; it needs 60 senators. Polymarket prices the bill being signed into law in 2026 at about 18%, down from 90% in February
Tue–Wed, Sep 15–16	FOMC meeting; rate decision Wednesday 2 p.m.	Odds of an increase were quoted between 48% and above 60% across venues on Tuesday; the target range is 3.50%–3.75%
Wed, Sep 16, 8:30 a.m.	August retail sales	Lands the morning of the FOMC decision; the consumption side of the same inflation question
Fri, Sep 18	September options expiry; Bank of Japan rate decision	The expiry’s notional size was still not reported when this brief was written; DWF Labs’ max-pain estimate of $78,000, dated September 4, is now two weeks old
Sat, Sep 19, ~12:52 p.m. UTC	Next Bitcoin difficulty adjustment, estimated +2.55% to about 130.7T	A measured retarget, rather than an inferred hashrate reading, is the network’s own answer to last week’s estimation noise
Wed, Sep 30, 8:30 a.m.	August PCE inflation and the third estimate of Q2 GDP	The Fed’s preferred inflation measure, two weeks after the decision it will not have informed
Through September	Calendar-month seasonality	September has finished negative in nine of sixteen completed years in this series and is its weakest month by average return, at −3.3% — though the last three Septembers closed positive

Reading the combination

The most useful thing about Tuesday is that it separates a signal from its story. A golden cross is, in the coverage, an event: it has a date, a name and a set of historical analogues that get attached to it within hours. In the data it is a threshold quietly crossed by two slow averages — the 50-day is the mean of closes since July 21, the 200-day the mean of closes since late February — and what crossed on September 8 was a statement about the summer. Price fell 0.82% the same day, closed at its lowest since September 2, and finished $18 from its own 20-day average, which is as close to directionless as a market gets. Both facts are true and they are not in tension; they are simply measured at different horizons. This series’ own record supports the deflationary reading: of the twelve golden crosses since 2012, price was higher thirty sessions later six times out of twelve, with a median of −2.6% — a coin flip — and only at sixty and ninety sessions does a positive tilt appear (eight of twelve, medians +12.6% and +10.7%). CoinDesk’s independent tally reaches the same conclusion from the other side, counting three of twelve signals that held for a year against a majority that produced what it calls premature bull traps. The layers otherwise disagree in a familiar pattern. Against: the MACD histogram widened negative for a fifth straight day to −396.55; ADX turned down for the first time in twenty-one sessions, confirming that the directional spread has been narrowing for two; the ten-day band-width contraction base rate (268 instances, 49% higher ten sessions later, median −0.3%) offers no edge; and September remains the weakest month in this record. For: 510 prior days with band width at or below 9% and price above its 200-day average were higher ten sessions later 62% of the time (median +1.4%) and twenty sessions later 65% (median +4.2%); 253 days with ADX at or above 45, +DI above −DI and ADX falling were higher 71% of the time at ten sessions (median +4.7%); volume below 0.70 times its average above the 200-day was higher 57% of the time; and $3.8 billion of ETF inflows over three weeks is real demand whatever the price has done with it. Three things are worth holding separately. First, the squeeze is still substantially mechanical rather than meteorological: band width fell 4.84 points because August 19’s $69,300 close left the twenty-day window, while thirty-day realized volatility rose to 47.25% — and one more such close, August 20’s $73,012, leaves after Wednesday, which on a flat session would take band width to roughly 5.6%. Anyone reading chart 3 as a market going quiet is reading a calendar effect. Second, Tuesday’s move was macro-shaped: equities, gold and Bitcoin all fell while oil rose 1.9% toward a three-month high, on a day when the odds of a Federal Reserve rate increase next week were being quoted anywhere between 48% and above 60% depending on the venue. That is not a crypto story. Third, the level structure is now the tightest of this entire run: $79,477 up to $80,000 immediately above, covering Tuesday’s high and the round number given back for a second session, with the upper band at $81,739 and CobraTrader’s $82,240 beyond; and $78,300 down to $77,589 below, covering Rekt Capital’s retest zone and Tuesday’s low, with September 2’s $76,219 as the month’s floor beneath that. The immediate bracket, from Tuesday’s low at $77,589 to the round number at $80,000, runs $2,411, about 1.06 times a single day’s ATR — and the narrower span between Rekt Capital’s $78,300 and that same round number is $1,700, or 0.75 times ATR. Either way, one ordinary session can resolve it. Which side it resolves toward is not something these indicators can tell you.

Sources

Price, volume and all technical indicators: Alpha Vantage BTC/USD daily series (00:00 UTC candles), computed by the author; the September 8 candle was confirmed complete by the presence of a partial September 9 row. Fear & Greed: the dated series from alternative.me’s API. ETF flows: Farside Investors’ per-fund table. The golden cross and its historical record: CoinDesk on the crossing, with the three-month projection from Crypto Briefing and Cryptonomist. Rekt Capital’s $78,300 condition, the oil move and The Kobeissi Letter: Cointelegraph on the support test. CobraTrader’s and Jesse Olson’s levels: CoinTurk. Galaxy Digital’s 50-week average, the $82,283 rejections and the Polymarket CLARITY Act pricing: 24/7 Wall St. Funding, open interest, liquidations and the CoinStats dominance figure: CoinStats daily analysis, used with the timing caveat described above. Corporate treasuries and the Liquid recovery: Decrypt on Strategy’s STRC buyback, Decrypt on Strive, Cointelegraph on the Liquid restart and Decrypt on the outstanding $47 million. Short-term-holder unrealized gains: Cointelegraph. Network: CoinWarz difficulty. Equity, gold and oil closes: Yahoo Finance’s September 8 market wrap. Fed odds and the Warsh repricing: Yahoo Finance on CME FedWatch and Motley Fool’s September 8 crypto wrap, which also supplies the late-session Ether, Solana and XRP marks. The inflation-report framing: Decrypt on Wall Street stalling before Friday. Dominance and market value: tv-hub.org. Calendar: Kiplinger’s week-ahead economic calendar. News sentiment tally: Alpha Vantage NEWS_SENTIMENT feed for September 8.