Bitcoin Daily Brief — September 12, 2026: A $445 Saturday, 2026’s 4th-Narrowest Range, as ETF Outflows Fade to $13M

Bitcoin price on September 12, 2026: BTC closed at $77,263, up 0.07%, in a $445 range between a $77,495 high and a $77,050 low — the fourth-narrowest session of 2026 and, on volume of 1,669 BTC, barely a quarter of the twenty-day average, the quietest aftermath imaginable to Friday’s $3,822 inflation-day round trip, with the Federal Reserve’s decision now four sessions away. 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 11, 2026. About this series: Bitcoin Daily Brief.

Bitcoin price on September 12, 2026, at a glance

The charts

The 30-second read

Bitcoin closed Saturday at $77,263, up $54.30 or 0.07%, in a $445.44 range — the fourth-narrowest session of 2026’s 255 and just 0.189 times Friday’s average true range — on 1,669 BTC of volume, 0.26 times the twenty-day average, which is as complete a stillness as this market produces and, on the evidence below, almost entirely a function of the calendar rather than the market. Underneath, the layers moved as little as the price: the trend layer added a twenty-fifth consecutive close above the 200-day average and widened the golden-cross gap a fifth session to $814.62, the momentum layer widened the MACD histogram a ninth day to −759.80 but by only $20.78, the smallest increment of the run, and the flow layer recorded a fourth straight ETF outflow at −$13.2 million, a twentieth the size of Thursday’s. What did change sits outside the chart: alternative.me’s Fear & Greed index rebounded seven points to 63, CoinWarz raised its estimate for the September 19 difficulty retarget from +3.62% to +4.25%, and the FOMC decision that CME FedWatch prices at roughly 85% for a rate increase is now four sessions away, with the Senate’s CLARITY Act cloture vote the day before it.

What happened in the last 24 hours

• Price. Opened $77,208.55, ranged only between $77,049.56 and $77,495.00 and closed $77,262.85 (+0.07%, +$54.30) — a $445.44 span on volume of 1,669.18 BTC, 0.26 times the twenty-day average; the body measured $54.30, or 12.2% of the range, with a $232.15 upper wick (the thin line above the candle body marking where price traded but did not close) and a $158.99 lower wick.
• Flows. Friday’s print arrived at −$13.2 million on Farside Investors’ table, a fourth consecutive outflow but roughly a twentieth of Thursday’s −$282.7 million; Bitcoin News Digest reported BlackRock’s IBIT at −$19.2 million with VanEck’s HODL and MSBT taking modest inflows, and described the deceleration as “selling exhaustion.” The four-session run totals −$462.7 million. September 12 was a Saturday, so there is no US ETF session to report — this is an absent number rather than a pending one.
• Positioning. CoinStats’ September 12 snapshot, timed 00:08 UTC, put perpetual funding at 0.0029% per eight hours — about 3.2% annualized — down from the 0.0039% it showed a day earlier, with open interest at $51.53 billion, off 3.05% (−$1.62 billion) over thirty days. Liquidations over the trailing twenty-four hours totalled $104.73 million, split $58.22 million short (55.6%) and $46.51 million long (44.4%) — a far more balanced reading than Friday’s two irreconcilable figures.
• On-chain. CoinWarz put difficulty at 127.45T with 1,019 of the current cycle’s 2,016 blocks mined (50.55%), and raised its estimate for the next retarget to September 19 at 04:46 UTC at +4.25%, to about 132.86T — up from the +3.62% to 132.06T it projected a day earlier. Bitdeer was reported to have sold its entire weekly production of 293.2 BTC, about $22.6 million at Saturday’s close, leaving no corporate treasury holdings.
• Macro and policy. US markets were closed for the weekend, so Friday’s closes stand: the ten-year Treasury yield at 4.959% and the thirty-year at 5.346% on TheStreet’s marks, gold at $4,404.40 and WTI crude at $99.05. CME FedWatch odds of a 25 basis-point increase at Wednesday’s FOMC were reported at 85.4% by Bitcoin News Digest, and EY-Parthenon chief economist Greg Daco was reported to have shifted his forecast to a 25 basis-point increase. Revolut confirmed that customer Bitcoin transaction histories, passport scans and addresses were disclosed to a spoofed regulatory request that passed email authentication.

Section summaries

Price and trend

Saturday was the photographic negative of Friday: a $445.44 range, the fourth-narrowest of 2026’s 255 sessions by percentage of the close and the third-narrowest by dollars, against Friday’s $3,822.22, which ranked twenty-ninth widest. The body was $54.30 and the close finished 47.9% of the way up the range — almost exactly in the middle, which is what a session with no argument in it looks like. The long-frame structure carried on unchanged: a twenty-fifth consecutive close above the 200-day average, now 10.23% above it, and a golden-cross gap that widened a fifth session from $617.00 to $814.62. Details ↓

Momentum and volatility

Both momentum measures moved, and both moved barely: RSI rose 0.17 points to 55.05, a second consecutive gain, while the MACD histogram widened a ninth straight day to −759.80 — but by $20.78, the smallest of the nine increments and roughly a third of Friday’s $60.33, which was itself a third of Thursday’s. ADX eased a fifth session to 44.32, and the directional spread narrowed from 15.40 to 15.18 after one session of widening. Volatility split again, in the opposite direction from Friday: ATR fell $136.28 to $2,217.04, giving back more than Friday’s $112.99 rise, while band width edged up from 6.015% to 6.127%. Details ↓

Flows and positioning

The ETF channel produced a fourth consecutive outflow, but a very small one: −$13.2 million on Farside Investors’ table against Thursday’s −$282.7 million, with IBIT reported at −$19.2 million and two smaller funds taking inflows, which Bitcoin News Digest characterised as “selling exhaustion.” The four-session run totals −$462.7 million, and September’s eight reported sessions remain net positive at +$307.3 million. On the derivatives side the picture calmed as much as the price: funding eased from 0.0039% to 0.0029% per eight hours on CoinStats’ readings, open interest slipped to $51.53 billion, and liquidations of $104.73 million split near evenly between longs and shorts. Details ↓

On-chain and network

The network reading moved in the one direction it can on a quiet weekend: CoinWarz now estimates the September 19 difficulty retarget at +4.25%, to about 132.86T, up from the +3.62% to 132.06T it projected a day earlier, with 1,019 of 2,016 blocks mined at 50.55% of the cycle. A rising projection with half the cycle complete is the cleanest available evidence that hashrate has grown, and it firms up as the sample does. Bitdeer was reported to have sold its entire weekly production of 293.2 BTC, approximately $22.6 million at Saturday’s close, maintaining no treasury holdings at all. Details ↓

Sentiment

alternative.me’s index dated September 12 reads 63 (“Greed”), up seven points from Friday’s 56 — a second consecutive same-source comparison, and a rebound that came on a day price moved 0.07%, which is the clearest possible illustration of why these indices are not price trackers. CoinStats showed 57 on its own separately built index the same morning; both are reported. The Alpha Vantage news feed carried only 7 Bitcoin-tagged articles, scored zero bullish, two somewhat bullish, four neutral, one somewhat bearish and zero bearish. Published views split cleanly: NordFX’s analytical group named $79,500 as what a recovery would need and $76,000 as what would open $72,000, while Motley Fool’s Dominic Basulto argued September weakness historically precedes a fourth-quarter rally. Details ↓

Macro and cross-asset

There is no Saturday macro session, so the numbers that matter are Friday’s and the expectations built on them. August core CPI at 0.3% against a 0.2% forecast moved CME FedWatch odds of a 25 basis-point increase at Wednesday’s FOMC to roughly 85% to 86.5%, from about 60% to 69% before the release; Bitcoin News Digest quoted 85.4% on Saturday, so the repricing held over the weekend rather than fading. The ten-year Treasury yield closed Friday at 4.959% and CoinGabbar reported yields at three-year highs near 4.99% on Saturday. Ether outperformed again, at $2,513.18 and up 2.6% on CoinGabbar’s Saturday mark against Bitcoin’s 0.6% on the same page. Details ↓

Calendar

Sunday September 13: the weekly candle closes at 8 p.m. ET, the frame on which the 50-week averages are defined — the simple average computed here sits at $79,644.68 and the exponential average at $77,382.54, which Saturday’s close finished $119.69 beneath. Tuesday September 15: the Senate’s CLARITY Act cloture vote. September 15–16: the FOMC, with the decision Wednesday at 2 p.m. ET, a press conference at 2:30 and a fresh Summary of Economic Projections, alongside August retail sales that morning. September 18: the September options expiry and the Bank of Japan decision; around September 19, the difficulty retarget now estimated at +4.25%. Details ↓

Reading the combination

Saturday’s headline reading is that Bitcoin went almost perfectly still — and the most useful thing this brief can say about it is that the stillness is mostly a property of the calendar, not of the market. All eight of 2026’s sessions with a range under 1% of the close have been Saturdays, against a 28.6% weekend share of the year’s sessions; every such session in 2024 and 2025 also fell on a weekend, where in 2017 the weekend share of quiet days was 33%. Compression has migrated to the weekend as weekday volatility has risen, so “the market went quiet on a Saturday” is closer to a tautology than a signal. Measured against the right comparison group — the 502 earlier Saturdays in this series with price above its 200-day average and a full forward record — quiet Saturdays do retain a modest tilt, but the layers themselves are unchanged from Friday: trend improving, momentum still deteriorating though at a decelerating rate, flows negative but barely, macro hostile and now only four sessions from resolution. 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 12 close)	Zone	Change since Sep 11	What this zone has historically meant
Bollinger Bands (20, 2)	%B 0.22; band width 6.13%	Lower third of the band; width still near the floor	%B 0.20 → 0.22; width 6.02% → 6.13%	Band width rose 0.112 points to 6.127%, its first increase in two sessions, because the standard deviation of the last twenty closes rose from $1,182.17 to $1,203.74 while the 20-day average fell $23.33. That is a rise driven by the arithmetic of the window rather than by Saturday’s trading: the $445 range contributed almost nothing, but the twenty-day window now contains Friday’s violent session, and it will keep containing it for nineteen more days. %B, which measures where the close sits between the bands with 0 at the lower band and 1 at the upper, edged from 0.2034 to 0.2248: the close finished $1,082.45 above a lower band at $76,180.40 and $3,732.53 below an upper band at $80,995.38. The envelope widened $86.30 to $4,814.97. Of 359 earlier days in this series with %B between 0.15 and 0.30 and price above the 200-day average, the close was higher ten sessions later 55% of the time (median +0.95%) and twenty sessions later 53% (median +0.53%). Of 366 earlier days with band width below 8% and price above the 200-day, the figures are 61% (median +1.37%) and 65% (median +4.89%) — the standing caveat being that low-band-width days cluster in a handful of quiet stretches, so this is not 366 independent observations.
Price vs 20-day SMA	−1.69% below ($78,588)	Below, fourth consecutive close	−1.78% → −1.69%	The gap narrowed from $1,402.66 to $1,325.04, and for once both halves of that moved: the close rose $54.30 and the average fell $23.33, reversing Friday’s $7.71 rise. The arithmetic is the one this brief set out yesterday, now running the other way: a twenty-day average falls when the close entering its window is lower than the one leaving. Saturday’s $77,262.85 replaced August 23’s $77,729.36, a shortfall of $466.51 spread over twenty days, hence the $23.33. The window now runs from August 24 to September 12, and August 24’s $78,981.59 leaves next — a higher close than any recent session, so on an unchanged close the average would fall again, and by more. Of 940 earlier days in this series with the close below a falling 20-day average while price held above its 200-day, the close was higher ten sessions later 52% of the time (median +0.21%) and twenty sessions later 52% (median +0.39%) — four points under the all-day baseline of 56%, and materially weaker than the 318-instance below-a-rising-20-day case carried yesterday, which ran 53% and 53% with medians of +0.65% and +1.23%. The difference between the two is small enough to sit inside noise, and is reported because the condition genuinely changed today, not because the change is decisive.
50-day vs 200-day SMA	$70,908 vs $70,093 (gap +$814.62)	Golden cross — 50-day above, fifth day	Gap +$617 → +$815	The gap widened for a fifth consecutive session since the cross printed, by $197.62, as the 50-day rose $263.59 to $70,907.56 — a thirty-second straight daily rise — against the 200-day’s $65.97 to $70,092.94. The mechanism has not changed since the cross printed: the 50-day window is still shedding the low closes of late July, which sit far below anything the 200-day window is shedding, so the shorter average gains ground almost regardless of what the current session does. Saturday is the purest demonstration of that available — the day’s own close moved the market $54.30 and the gap still widened by nearly $200. The base rates are struck on the cross date rather than on subsequent days and are therefore unchanged: of the twelve earlier golden crosses this series records, beginning with February 9, 2012, price was higher thirty sessions later in six of twelve cases (median −2.62%), sixty sessions later in eight of twelve (median +12.60%) and ninety sessions later in eight of twelve (median +10.67%). Twelve observations is a sample in which one instance moves the median by several points, and that is how it should be read.
Price vs 200-day SMA	$77,263 vs $70,093 (+10.23%)	Above (25th straight close, since Aug 19)	+10.26% → +10.23%	Price above the 200-day average is the most common definition of a long-term uptrend, and this is a twenty-fifth consecutive close above it. The distance slipped 0.03 percentage points — not because price fell, but because the average rose $65.97 while the close rose only $54.30, which is the first time in this run that the line has gained on the price on a day when price rose. The cushion stands at $7,169.91; it would take a 9.28% decline from Saturday’s close to reach the line. The 200-day has now risen on each of the last twenty-two sessions. Its rate of increase is the reading to watch here rather than the distance: $65.97 on Saturday against $62.91 on Friday and $44.59 on Thursday, an acceleration that reflects the closes now entering its window being progressively higher relative to those leaving. A market that stood still for a day while its long average climbed $66 is a market whose trend layer is doing the work unaided.
RSI (14)	55.05	Mid neutral zone (30–70)	54.87 → 55.05	Up 0.17 points, a second consecutive rise after the four-session decline that ran −5.17, −2.55, −0.66 and −6.68. RSI compares the average size of up-days with down-days over fourteen sessions, so a +0.07% close barely registers — this is the smallest daily RSI move in the current sequence and it should be read as “unchanged” rather than as a second leg of recovery. The reading remains far below the 86 it printed on August 21 and inside the band this brief has called neutral for two weeks. Of 615 earlier days in this series with RSI between 52 and 58 and price above its 200-day average, the close was higher ten sessions later 51% of the time (median +0.19%) and twenty sessions later 49% (median −0.18%) — the flattest base rate on this board and, at twenty sessions, the only one below a coin flip. That is the honest reading of a mid-range RSI: it is the zone in which the indicator carries the least information, and a 0.17-point move does nothing to change that.
MACD (12, 26, 9)	1,811 vs signal 2,571; histogram −759.80	Below signal, still far above zero	Histogram −739.02 → −759.80 (widening)	A ninth day below the signal line and a ninth consecutive widening, the sequence now reading −49.50, −108.96, −139.48, −261.93, −396.55, −500.09, −678.69, −739.02, −759.80. The nine daily increments run $64.50, $59.46, $30.52, $122.45, $134.62, $103.54, $178.60, $60.33 and $20.78, so Saturday’s $20.78 is the smallest of them all — about a third of Friday’s, which was itself about a third of Thursday’s. Two consecutive sharp decelerations is the most substantive momentum development on this board, though it remains a deceleration of deterioration rather than a reversal of it: the histogram is still making new lows for the sequence. The MACD line fell $210.73 to 1,811.06 while the signal eased $189.95 to 2,570.86. At −759.80 the histogram is the most negative since June 10, but the MACD line remains 1,811 above zero, so this describes a loss of upward momentum rather than a change of trend direction. Of 631 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 +0.84%) and twenty sessions later 53% (median +0.80%).
ADX (14)	44.32 (+DI 28.7, −DI 13.5)	Strongly trending (above 25), falling for a fifth day	44.97 → 44.32	A fifth consecutive fall, of 0.64 points, and this time the internals fell with it rather than against it: +DI eased from 29.11 to 28.70 and −DI from 13.71 to 13.51, narrowing the directional spread slightly from 15.40 to 15.18 after Friday’s single widening. On a day with a $445 range that is close to mechanically inevitable — the directional movement indices are built from the day’s high-to-high and low-to-low extensions, and a session that barely moves supplies almost no directional movement in either direction, so both indices decay toward their smoothed averages. A falling ADX from a high level describes a strong trend losing intensity, not a trend reversing. Days in this series with ADX at or above 44, +DI above −DI and ADX falling (266 earlier instances, recomputed for today’s data) were followed by a higher close ten sessions later 70% of the time (median +4.36%) and twenty sessions later 68% (median +5.35%) — the strongest reading on this board, with the same caveat carried every day: such instances cluster in a handful of strongly trending years, so this is not 266 independent observations.
ATR (14)	$2,217 (2.9% of price); range $445.44, upper wick $232.15	Falling; gave back more than Friday’s rise	$2,353 → $2,217	Down $136.28, more than undoing Friday’s $112.99 increase, and the cause is the simplest on this board: a high-to-low range of $445.44, which is 0.189 times Friday’s ATR and 0.577% of the closing price. That ranks fourth-narrowest of 2026’s 255 sessions by percentage of the close and third-narrowest by dollars — behind only August 15’s $293.67 and August 8’s $360.27 on the dollar measure, both of which were also Saturdays, with January 10’s $446.57 a further $1.13 wider than Saturday’s. ATR is a fourteen-day Wilder average, so one very quiet session pulls it down by roughly a fourteenth of the gap between the day’s true range and the prior average, which is what happened. The body was $54.30, or 12.19% of the range; the upper wick $232.15, the lower $158.99, and the close sat 47.88% of the way up from the low. Of 692 earlier days in this series with ATR falling after a rise while price held above its 200-day average, the close was higher ten sessions later 57% of the time (median +1.45%) and twenty sessions later 59% (median +2.93%) — a shade above the all-day baseline of 56%, and one of several readings on this board that lean mildly constructive without saying anything strong.
30-day realized volatility	47.6% annualized	Elevated; near the highest since April	47.59% → 47.58%	Down 0.01 points — unchanged to any practical precision, for a second consecutive session. Realized volatility measures how much daily returns have varied over thirty days, and Saturday’s return of +0.07% is small enough to be near the centre of any recent distribution, so it neither raised nor lowered the dispersion. The measure has now spent six sessions between 46.75% and 47.67%, its highest stretch since early April. The point worth carrying from Friday still applies and is now doubly demonstrated: band width and realized volatility both use closing prices, so neither registered Friday’s $3,822 intraday round trip as volatile; on Saturday, by symmetry, neither registered a $445 range as calm. ATR is the only measure on this board that sees intraday range at all, and it is the only one that moved meaningfully on either day — up $113 on Friday, down $136 on Saturday. When three volatility measures disagree, the question to ask is which of them is allowed to see the thing you care about.
Volume vs 20-day average	1,669 BTC vs 6,392 average (0.26×)	Far below the average — ninth-lowest ratio of 2026	7,311 → 1,669 BTC	Down 77.2% from Friday and the lightest session since August 15, with the ratio to the twenty-day average at 0.261 — the ninth-lowest of 2026’s 255 sessions, against January 17’s 0.207 low. The twenty-day average itself fell from 6,539 to 6,392 because the session entering the window was far lighter than the one leaving it. Context matters more than the rank here: every one of 2026’s ten lightest sessions by this ratio falls on a Saturday or Sunday, and the weekend accounts for 28.6% of the year’s sessions, so a light Saturday is the ordinary state of the world rather than an event. The related base rate is thin and should be treated as such: of only 58 earlier days in this series with volume below 0.35 times its twenty-day average while price held above the 200-day, the close was higher ten sessions later 60% of the time (median +1.64%) and twenty sessions later 57% (median +1.34%). Fifty-eight observations across a fourteen-year record is too few to carry weight, and the direction of the tilt is well inside what a sample that size produces by chance.
US spot ETF net flows	Sep 11: −$13.2M; four-day −$462.7M; September ≈ +$307.3M; August +$3,524M; no Sep 12 session	Fourth consecutive outflow, but the smallest of the four	−$282.7M → −$13.2M	Friday’s print closed the gap this brief flagged as pending yesterday, and it came in at −$13.2 million on Farside Investors’ table — a fourth consecutive redemption day, but roughly a twentieth of Thursday’s −$282.7 million and small enough that it is better read as flat than as an outflow. Bitcoin News Digest reported BlackRock’s IBIT at −$19.2 million, with VanEck’s HODL and MSBT taking modest inflows, which means the headline figure is a single large fund’s redemption partly offset elsewhere rather than a broad withdrawal; the same report described the deceleration as “selling exhaustion,” which is that publication’s characterisation and not a reading this brief endorses or tests. The four-session run now totals −$462.7 million against a three-week inflow run into September 8 of roughly $3.8 billion, so the pullback has given back about 12% of it. September’s eight reported sessions remain net positive at about +$307.3 million, and August closed at +$3,524 million. There is no September 12 figure because September 12 was a Saturday and US ETFs did not trade — an absent number, not a pending one, and the distinction matters when reading the chart’s rightmost bar.
Perpetual funding rate	+0.0029% per 8 hours (≈ +3.2% annualized)	Mildly positive; longs paying shorts	0.0039% → 0.0029%, first same-source comparison	Funding is the periodic payment between holders of perpetual futures that keeps the contract tethered to spot; positive means longs pay shorts, which indicates net long positioning. CoinStats’ September 12 read, timed 00:08 UTC, put it at 0.0029% per eight-hour period, which annualizes to roughly 3.2% (0.0029 × 3 × 365). This is the first day-over-day funding comparison this brief has been able to make, both readings coming from CoinStats, and it shows a mild long bias getting milder — down about a quarter in a session. The level remains far below the crowded-long readings that have preceded sharp unwinds in this series, which have generally run several times higher. One caveat on the comparison: both figures are single snapshots from one provider taken at similar times of day, not a time-weighted average across venues, so the direction is more reliable than the magnitude of the change.
Open interest / liquidations	OI $51.53B; liquidations $104.73M (55.6% shorts, 44.4% longs)	OI down; liquidations small and near-balanced	OI $53.17B → $51.53B (same source)	Liquidations are forced closures of leveraged positions by an exchange when margin runs out. After Friday’s two irreconcilable readings — $427 million of shorts on Bloomberg’s data against $75.72 million of 92.9% longs on CoinStats’ — Saturday produced a single, small, near-balanced figure: $104.73 million in twenty-four hours, $58.22 million short (55.6%) and $46.51 million long (44.4%). That is the shape of a market with no dominant one-sided position being forced out, and it is consistent with the near-absence of price movement. Open interest fell from $53.17 billion to $51.53 billion, a $1.64 billion decline; because both figures come from CoinStats this is the first genuine day-over-day open-interest comparison this brief has been able to make in ten editions, and it shows leverage coming out of the market rather than building — CoinStats reports the thirty-day change at −3.05%, having shown +13.67% a day earlier, which is a large swing in a thirty-day window and is quoted as the provider states it. Declining open interest alongside a flat price describes positions being closed rather than reversed.
Fear & Greed	63 (“Greed”)	Greed, mid-band	56 → 63, second consecutive same-source comparison	alternative.me’s dated API returned data for a second consecutive run, extending the series: 69 on September 8, 66 on September 9, 69 on September 10, 56 on September 11 and 63 on September 12. The seven-point rebound recovers just over half of Friday’s thirteen-point fall — which was the largest single-day move this brief has recorded — and it happened on a day Bitcoin moved 0.07%. That pairing is the whole argument against reading this index as a sentiment thermometer: it blends volatility, momentum, dominance and survey inputs rather than tracking price, so Friday’s burst of intraday range pushed it down and Saturday’s stillness pushed it back up, both times independently of what price did. CoinGabbar reported the same 63 for September 12, which is consistent because it carries alternative.me’s index; CoinStats’ separately constructed reading showed 57 the same morning, and both are reported as distinct measurements rather than reconciled.
Hashrate / difficulty	Difficulty 127.45T; next retarget estimated Sep 19, +4.25% to ~132.86T	Between retargets, 50.55% through the cycle	Estimate +3.62% → +4.25%	CoinWarz shows difficulty unchanged at 127.45T — it changes only every 2,016 blocks — with 1,019 blocks of the current cycle mined, 50.55% complete, and now estimates the next retarget for September 19 at 04:46 UTC at +4.25%, taking difficulty to about 132.86T. Both the size and the timing moved from yesterday’s +3.62% to 132.06T at 08:54 UTC, and both moved in the same direction for the same reason: blocks are arriving faster than the ten-minute target, which simultaneously raises the projected adjustment and pulls the retarget date earlier. Difficulty is the protocol’s self-correcting mechanism, adjusting every fortnight so that block intervals stay near ten minutes, so a projected +4.25% means the current cycle’s blocks are arriving about 4.25% ahead of schedule. The estimate is now struck on 1,019 actual blocks rather than 904, which makes it materially more reliable than yesterday’s, and it will keep firming until the retarget settles it on the full 2,016. Spot hashrate figures remain unquotable for the reason carried all week — they are inferred from block intervals rather than measured, and short-window estimates from the same sources scatter by roughly a tenth — but this rising difficulty projection is the indirect evidence that the underlying quantity has grown.

Price and trend (chart 1)

Saturday is most easily described by what did not happen. Bitcoin opened at $77,208.55 — Friday’s close, as the daily candles here are continuous — traded down to $77,049.56 and up to $77,495.00, and closed at $77,262.85, a gain of $54.30 or 0.070%. The range was $445.44, which is 0.577% of the closing price and 0.189 times Friday’s ATR of $2,353.32. That ranks fourth-narrowest of 2026’s 255 sessions by percentage of the close and third-narrowest by dollars. To put the two days side by side: Friday’s range was 8.6 times Saturday’s, and Friday’s upper wick alone, at $2,643.67, was nearly six times Saturday’s entire high-to-low span. The body — the distance between open and close — was $54.30, or 12.19% of the range, with a $232.15 upper wick and a $158.99 lower wick, and the close finished 47.88% of the way up from the low. A candle with a tiny body sitting in the middle of a tiny range, with near-symmetric wicks, is the chart’s way of recording an absence of argument. Volume confirms it: 1,669.18 BTC against a twenty-day average of 6,392.43, a ratio of 0.261 and the lightest session since August 15.

The level structure beneath the market repaired itself simply by the market not going there. Friday’s low of $76,030.00 remains the September low, and it remains the only session low this month to have pierced the lower Bollinger Band. Saturday’s low of $77,049.56 finished $869.16 above that band, which now sits at $76,180.40 — so the cluster of reference points that Friday tested and closed above was not revisited at all. Against longer reference points the close sits 38.82% below the all-time high of $126,296, 11.70% below the 2025 year-end close of $87,497.94, and 32.02% above the June 30 low close of $58,523.93 — or, stated the other way round so the direction is unambiguous, that June low is 24.25% below Saturday’s close. September is 1.65% lower across its first twelve sessions, a $1,299.89 decline from August 31’s $78,562.74, and the close sits 19.72% of the way up the month’s range between the $76,030.00 low and the $82,283.00 high of September 3. This record contains sixteen completed Septembers; nine finished negative, and the month’s average return of −3.3% is the weakest of the twelve, though the last three Septembers closed +4.0%, +7.3% and +5.4%.

The long averages did almost all of the day’s work. The 50-day rose $263.59 to $70,907.56 — a thirty-second consecutive daily rise — while the 200-day rose $65.97 to $70,092.94, taking the golden-cross gap from $617.00 to $814.62, a widening of $197.62 and the fifth consecutive session of widening since the cross printed on September 8. Set that against the day’s own price move of $54.30 and the point makes itself: the gap between the two averages widened by nearly four times what the market actually moved. Bitcoin closed 10.23% above the 200-day average, 8.96% above the 50-day, 15.31% above the 100-day at $67,006.08 and 1.69% below the 20-day at $78,587.89 — a twenty-fifth consecutive close above the 200-day in the run that began on August 19, and a fourth below the 20-day. One detail is new and runs against the run’s direction: the distance to the 200-day narrowed very slightly, from 10.26% to 10.23%, for the first time in this sequence on a day when price rose. The reason is that the 200-day gained $65.97 while the close gained $54.30, so the line closed $11.67 of ground. The cushion is $7,169.91 and it would take a 9.28% decline to reach the line, so nothing about this is precarious — but it is the first sign that a rising long average eventually catches a market that stops rising, which is the mechanical counterpart to the trend layer’s otherwise unbroken improvement.

Levels being discussed (attributed)

Level	Why it is being watched	Who cites it
$126,000	The end-2026 figure reached by applying October’s +20% and November’s +40% historical averages to an assumed $75,000 September trough — equal to the October 2025 all-time high	Dominic Basulto, Motley Fool (Sep 12)
$86,000	Upper bound of the stated base-case trading range for the week of Sep 14–18	NordFX Analytical Group (Sep 12)
$82,833 / $82,281	Cited as the prior highs overhead	Jamie Redman, Bitcoin.com News (Sep 12)
$82,283	The September 3 high and the month’s high, the highest print since May 11	Alpha Vantage feed
$81,405	The 365-day moving average, computed here	This brief’s calculation
$80,995	Upper Bollinger Band, 4.8% above the close and $20 higher than yesterday	This brief’s calculation
$80,000–$82,000	Described as the resistance zone overhead	CoinStats AI market page (Sep 12)
$79,668	The 50-week moving average, described as capping a short-squeeze attempt on Saturday	Bitcoin News Digest (Sep 12)
$79,645	50-week simple moving average, computed here on weekly bars; the close is $2,382 below it	This brief’s calculation
$79,500	Named as the level a recovery would need to clear to revive the prior uptrend	NordFX Analytical Group (Sep 12)
$78,800–$77,970	Cited as the near-term resistance band	Jamie Redman, Bitcoin.com News (Sep 12)
$78,588	20-day simple moving average, $1,325 above the close and falling again after one day’s rise	This brief’s calculation
$77,971	The 10-period exponential moving average, cited as leaning bearish on the short term	Jamie Redman, Bitcoin.com News (Sep 12)
$77,495	Saturday’s high on the 00:00 UTC candle	Alpha Vantage feed
$77,383	The 50-week exponential moving average — the “line in the sand” named earlier in the week — computed here on weekly bars; Saturday’s close finished $120 beneath it	Level named by Ted Pillows via Cryptonews (Sep 11); figure computed here
$77,345	The price quoted in Saturday-afternoon coverage at 16:38 UTC, down 0.70% on that feed’s rolling day	ts2.tech (Sep 12)
$77,263	Saturday’s close, 19.7% of the way up the September range between the month’s $82,283 high and its $76,030 low	Alpha Vantage feed
$77,050	Saturday’s low on the 00:00 UTC candle — $1,020 above Friday’s low	Alpha Vantage feed
$77,000–$77,200	Described as the support the pair was trading on at the end of the week	NordFX Analytical Group (Sep 12)
$76,850	Named as “the first evidence that buyers can defend the current band”; it is the low of that publication’s rolling 24-hour window at 16:38 UTC Saturday, which overlaps Friday and sits below the 00:00 UTC candle’s low	ts2.tech (Sep 12)
$76,400–$76,040	Cited as the near-term support band	Jamie Redman, Bitcoin.com News (Sep 12)
$76,180	Lower Bollinger Band, $1,082 below the close and $66 lower than yesterday; Saturday’s low stayed $869 above it	This brief’s calculation
$76,030	Friday’s low — the September low, and described as “the stronger line” beneath the market	Alpha Vantage feed; framing from ts2.tech (Sep 12)
$76,000	Named as the level whose breach would risk an extension toward $72,000	NordFX Analytical Group (Sep 12)
$72,941	200-day exponential moving average	This brief’s calculation
$70,908 / $70,093	50-day and 200-day simple moving averages — the two lines that crossed on September 8, now $815 apart	This brief’s calculation
$69,500	Lower bound of the stated base-case trading range for the week of Sep 14–18	NordFX Analytical Group (Sep 12)
$58,524	The June 30 low close of 2026 — 24.2% below Saturday’s close, which sits 32.0% above it	Alpha Vantage feed

One note on the table. Of these twenty-eight entries, fourteen carry a September 12 attribution to published coverage, seven are this brief’s own calculations, five come straight off the price feed, and two are hybrids — the 50-week exponential average, a level named by an analyst earlier in the week whose value is computed here, and Friday’s $76,030 low, a feed number carrying a September 12 framing. The structural feature of Saturday’s version is how crowded the area immediately around the close has become. Within $250 of the $77,263 close sit the session high $232 above, the 50-week exponential average $120 above, the ts2.tech Saturday mark $82 above, the session low $213 below and the top of NordFX’s $77,000–$77,200 support band $63 below; the 10-period exponential average adds a sixth point $708 above. That is five reference points inside a span of $445 — the width of Saturday’s entire range — and six inside $971 — the mirror image of Friday, when a $3,822 candle swallowed the entire nearby level structure. Stripping out the long-horizon projections, the working band named in Saturday’s published analysis runs from $69,500 to $86,000, NordFX’s stated base case for the coming week, which is $16,500 wide, or about 7.44 times Saturday’s ATR of $2,217.04. The Bollinger envelope is far tighter at $76,180 to $80,995, a $4,814.97 span against $4,728.67 a day ago; the envelope widened by $86.30, and it did so by dropping its floor $66.48 and lifting its ceiling $19.83.

Momentum and volatility (charts 2 and 3)

The momentum measures agreed on direction for the first time in three sessions, and both agreed on very little movement. RSI rose 0.17 points to 55.05, a second consecutive gain, which on a +0.07% close is about as small as the indicator can register while still moving. MACD widened for a ninth consecutive session: the line fell $210.73 to 1,811.06 while the signal eased $189.95 to 2,570.86, taking the histogram from −739.02 to −759.80. The detail that matters is the rate. The nine daily widening increments run $64.50, $59.46, $30.52, $122.45, $134.62, $103.54, $178.60, $60.33 and $20.78, so Saturday’s is the smallest of the nine — and the sequence of the last three, $178.60 then $60.33 then $20.78, is two consecutive reductions of roughly two-thirds each. The MACD line keeps falling because the twelve-day exponential average is still shedding the memory of September 3’s $81,263.99 close faster than the twenty-six-day average is, and a flat session does not arrest that; but the gap it is opening per day has shrunk by an order of magnitude in three sessions. This is a deceleration, not a turn — the histogram is still setting new lows for the run —. ADX fell a fifth straight session, to 44.32 from 44.97, and this time the internals moved with it: +DI eased from 29.11 to 28.70 and −DI from 13.71 to 13.51, narrowing the directional spread from 15.40 to 15.18 after Friday’s single widening. On a session with a $445 range that is close to mechanically forced, since the directional movement indices are constructed from daily high and low extensions and a day that barely extends in either direction feeds almost nothing into them.

The volatility measures split again, and in exactly the opposite configuration from Friday, which is the clearest demonstration of their construction this brief has been able to offer. ATR fell $136.28 to $2,217.04, giving back more than Friday’s $112.99 rise, because ATR uses the session high and low and Saturday’s $445.44 range was 0.189 times the prior day’s average. Bollinger band width, computed from the standard deviation of the last twenty closes, went the other way: it rose from 6.0153% to 6.1269% as the standard deviation increased from $1,182.17 to $1,203.74 and the 20-day average fell $23.33. Note carefully what caused that rise — not Saturday’s trading, which contributed a close within $54 of Friday’s, but the fact that the twenty-close window now contains Friday’s $77,208.55 alongside the early-September closes above $80,000, and dispersion in that window is what band width measures. Thirty-day realized volatility, also computed from closes, was unchanged to two decimal places at 47.58% against 47.59%. So on Friday only ATR saw a volatile day; on Saturday only ATR saw a calm one. Neither result is a flaw in any of the three measures. It is a reminder that “volatility” is not one quantity: a closing-price measure and a range measure can move in opposite directions on the same session indefinitely, and a reader who tracks only one will periodically be told the opposite of what a reader tracking the other is told.

The base rate worth testing on a day like this is the intuition that a very quiet session is a coiled spring. The straightforward version of that test is contaminated, and the contamination is the finding. Take every earlier session in this series with a range of 0.6% of the close or less while price held above its 200-day average: 704 instances, followed by a higher close ten sessions later 64% of the time (median +3.08%) and twenty sessions later 63% (median +6.92%), against baselines of 58% and 58% for days above the 200-day. That looks like a strong result. But the filter is partly selecting for the day of the week rather than for market compression, and the degree to which that is true has changed dramatically over this record. In 2017, 33% of the sessions with a sub-1% range fell on a weekend; in 2018, 42%; in 2020, 55%; in 2023, 81%; and in each of 2024, 2025 and 2026, every single one — eight of eight so far this year — against a weekend share of 28.6% of all sessions. As weekday volatility has risen, the only days quiet enough to clear a low-range filter are the days the traditional-finance world is shut. So the honest comparison is Saturday against Saturdays. Of the 502 earlier Saturdays in this series that had price above the 200-day average and a full twenty-session forward record, the close was higher five sessions later 57% of the time (median +0.80%), ten sessions later 59% (median +1.49%) and twenty sessions later 57% (median +2.79%). Narrowing to the 107 of those that also had a sub-0.6% range gives 66% (median +2.48%), 64% (median +2.90%) and 61% (median +5.73%). The tilt survives the correction, but it is roughly half the size the uncorrected figure implied, and 107 instances spread over fourteen years is a modest sample. The separate question — what follows stillness that arrives immediately after violence, as this did — cannot be answered here at all: the filter of a sub-1% range on the day after a 4%-plus range while above the 200-day returns seven earlier instances, which is too few to quote a percentage from, and this brief will not manufacture one.

Flows and positioning (chart 4)

Chart 4’s newest bar is barely visible, and that is the news. Friday’s US spot Bitcoin ETF net flow — the figure this brief listed as pending yesterday — printed at −$13.2 million on Farside Investors’ table. It is a fourth consecutive outflow, which continues the run, but at roughly one-twentieth of Thursday’s −$282.7 million it is small enough that “flat” describes it better than “outflow”: on a $97.5 billion asset base it is a rounding error of about 0.014%. Bitcoin News Digest reported the composition, and the composition is the interesting part: BlackRock’s IBIT shed $19.2 million while VanEck’s HODL and MSBT took modest inflows, so the net figure is one fund’s redemption partially offset by others rather than a broad withdrawal across the complex. That publication described the pattern as “selling exhaustion” — their characterisation, reported here as an attributed view rather than as a conclusion this brief has tested; four days is not a sample from which exhaustion can be established. The four-session run totals −$462.7 million, against a three-week inflow run into September 8 of roughly $3.8 billion, so the pullback has given back about 12% of it. September’s eight reported sessions still stand at about +$307.3 million against August’s completed +$3,524 million. There is no September 12 print, and the reason is worth stating plainly because it differs from the usual one: September 12 was a Saturday, so the US ETF market did not open. Where this brief normally flags the latest session’s flow as “not yet reported,” here the number does not exist and never will. Separately, ts2.tech noted that IBIT closed Friday at a $43.71 net asset value against a $43.77 market close, and that at Saturday-afternoon Bitcoin prices it put the indicative NAV at about $43.79, which it described as only 0.04% above that closing share price — a tight tracking figure, and the mechanism by which the weekend gap between a 24/7 asset and a weekday-traded fund is normally closed at Monday’s open.

The derivatives picture calmed as comprehensively as the price, and for the first time in this series both of its headline numbers can be compared with the day before from the same provider. CoinStats’ September 12 page, timed 00:08 UTC, put perpetual funding at 0.0029% per eight-hour period — roughly 3.2% annualized — against 0.0039% on its September 11 page, a decline of about a quarter. Open interest fell from $53.17 billion to $51.53 billion, a $1.64 billion reduction; because both readings are CoinStats’, this is a genuine day-over-day comparison rather than the source-change caveat this brief has carried for nine consecutive editions. The same page reports the thirty-day open-interest change at −3.05%, where a day earlier it showed +13.67% — a swing far larger than one session’s $1.64 billion can explain, so it is quoted as the provider states it and not treated as a measured change. Liquidations were small and near-balanced: $104.73 million across twenty-four hours, $58.22 million of shorts (55.6%) and $46.51 million of longs (44.4%). Set that against Friday’s problem — Bloomberg reporting $427 million of short liquidations while CoinStats showed $75.72 million skewed 92.9% long — and Saturday’s single, modest, two-sided figure is what an uncontested market looks like in the derivatives data. Falling open interest alongside a flat price and balanced liquidations describes leverage being closed out rather than flipped from one side to the other. On the wider market, CoinGabbar put Bitcoin’s share of total crypto value at 56.8% with Ether at 11.2% and total market capitalization at $2.73 trillion, up 0.8% on the day; Ether itself was quoted at $2,513.18, up 2.6% against Bitcoin’s 0.6% on that page’s own rolling day — a third consecutive session of Ether outperformance on these marks.

On-chain and network

The network reading is the one number on this page that moved decisively, and it moved in a way that is only visible day over day. CoinWarz shows difficulty at 127.45T — unchanged, as it must be between retargets — with the chain 1,019 blocks into the current 2,016-block cycle, 50.55% complete, and now projects the next retarget for September 19 at 04:46 UTC at +4.25%, taking difficulty to about 132.86T. Yesterday the same source projected +3.62% to 132.06T at 08:54 UTC. Both the size of the adjustment and its timing moved, and they moved together for one reason: blocks are arriving faster than the ten-minute target, which simultaneously raises the projected increase and pulls the estimated retarget time earlier. Difficulty is the protocol’s self-correcting mechanism — every 2,016 blocks, roughly a fortnight, it adjusts so that blocks keep arriving about every ten minutes — so a projected +4.25% means the current cycle is running about 4.25% ahead of schedule, a larger overshoot than the +1.31% the previous cycle produced and a larger one than yesterday’s estimate implied. The estimate’s reliability matters as much as its value: it is now struck on 1,019 observed blocks against 904 yesterday, so more than half the cycle is in the sample. This is the cleanest available evidence that hashrate has grown over the past fortnight, and it is more trustworthy than any of the spot hashrate figures, which this brief continues to decline to quote because estimates from the same sources scatter by roughly a tenth across the same moment. The September 19 retarget will settle the question with a figure derived from 2,016 actual blocks rather than an extrapolation.

The corporate and security items are both about custody, from opposite ends. Bitcoin News Digest reported that Bitdeer liquidated its entire weekly production of 293.2 BTC, approximately $22.6 million at Saturday’s closing price, and that the company holds no Bitcoin in treasury at all. That is a policy rather than a market signal — a miner selling production as it is mined converts a volatile asset into operating cash and removes price exposure from the balance sheet — but it sits interestingly alongside the treasury-company debate: Motley Fool’s Dominic Basulto published a piece the same day arguing against Bitcoin treasury companies generally, on the grounds that Strategy sold Bitcoin near $60,000 and bought near $80,000, which he read as evidence that such vehicles do not reliably add value over holding the asset. A miner with zero treasury and a treasury company with a large one are the two poles of that argument, and both were in Saturday’s coverage. At the other end, Decrypt and Cointelegraph reported that Revolut confirmed the unauthorised disclosure of customer data — Bitcoin transaction histories, passport scans and addresses for a “limited” number of users — after fulfilling a fraudulent information request that was sent from a government agency’s own email domain and passed email authentication checks. The relevance to this brief is narrow but real: it concerns the custodial and identity layer that sits between users and the protocol, not the protocol itself, which is the same distinction this brief drew about the Liquid sidechain exploit earlier in the week. No on-chain valuation metrics — MVRV, exchange balances, SOPR — were sourced on this run; the free-tier providers that carry them publish on a weekday schedule and had nothing dated September 12 when this brief was written.

Sentiment

alternative.me’s dated API returned data for a second consecutive run, which now gives this brief a continuous five-day series: 69 on September 8, 66 on September 9, 69 on September 10, 56 on September 11 and 63 on September 12, all inside the “Greed” band, which begins at 55. Saturday’s seven-point rebound recovers just over half of Friday’s thirteen-point fall, and the circumstances of the two moves make the same point from opposite directions. Friday’s collapse came on a day Bitcoin rose 0.88%; Saturday’s recovery came on a day Bitcoin rose 0.07% and traded in a $445 band. Neither move tracked price, because the index does not track price — it blends volatility, momentum, market dominance and survey components, so the burst of intraday range on Friday pushed it down and the return to stillness on Saturday pushed it back up. A reader who took 63 as a statement that investors turned more confident over the weekend would have the causation backwards. CoinGabbar reported the same figure of 63 for September 12, which is consistency rather than corroboration since that publication carries alternative.me’s index; CoinStats’ separately constructed reading showed 57 on the same morning, and the six-point spread between two indices measuring nominally the same thing is the ordinary state of this data.

Of the 7 Bitcoin-tagged articles in the Alpha Vantage news feed published between 00:00 UTC on September 12 and 04:00 UTC on September 13 — down sharply from 24 on Friday — the feed’s sentiment model scored none as bullish, two as somewhat bullish, four as neutral, one as somewhat bearish and none as bearish. Seven articles is the thinnest tally this brief has recorded, and the reason is the same as everything else on this page: Saturday. What little there was skewed away from the market itself. The somewhat-bullish scores went to a Motley Fool piece on September seasonality and to a Cointelegraph report that Nvidia is weighing a $10 billion investment in Anthropic’s IPO, which touches Bitcoin only through the mining-adjacent equities in the same basket. The single somewhat-bearish score attached to Decrypt’s report on the Revolut data disclosure — a security story, not a market one. The remaining four were neutral, including a Cointelegraph report on North Korean IT-worker infiltration and a Motley Fool piece on XRP. As always, a tally like this measures the tone of coverage as a machine reads it across everything tagged to the ticker, and on a seven-article Saturday it is reported for continuity rather than as a market indicator; a sample this small should not be compared with Friday’s twenty-four as though the two were measuring the same population.

Published views, in both directions. The most specific near-term framing came from NordFX’s analytical group, writing its outlook for the week of September 14–18: Bitcoin fell more than 5% on the week amid what it called “hawkish Fed repricing,” it described momentum indicators as sitting “near neutral to bearish with no strong directional signal,” and it named the levels on both sides — a move above $79,500 as what would be “needed to revive the bullish trend,” and a break below $76,000 as what “risks an extension toward $72,000,” inside a stated base-case range of $69,500 to $86,000. The constructive case was seasonal and came from Motley Fool’s Dominic Basulto, who noted that September is historically Bitcoin’s weakest month at an average −2.93% on his figures, and set against it October, November and December averages of +20%, +41% and +4%, sketching a path in which a September trough near $75,000 followed by average October and November returns would reach about $126,000 — which he observed is the October 2025 all-time high “on the dot.” That worked path reaches the figure using a 40% November leg rather than the 41% average quoted alongside it; at 41% it lands nearer $126,900. That is an arithmetic illustration of historical averages rather than a forecast, and the averages behind it are drawn from a small number of years in which a handful of extraordinary months dominate. The same author published a second piece the same day arguing the bearish side of a different question, that Bitcoin treasury companies are “a clear stay-away” because Strategy sold near $60,000 and bought near $80,000. Bitcoin.com News’ Jamie Redman, surveying the technical picture, counted one bearish, eight neutral and two bullish oscillator readings and ten bullish against four bearish moving-average readings, summarising it as “Bitcoin’s bigger trend is still standing while the short-term machinery works through a rough patch” — which is, in plainer words, the same split between the trend layer and the momentum layer that this brief has described all week. ts2.tech named $76,850 as “the first evidence that buyers can defend the current band” and $76,030 as “the stronger line,” adding that a breach of the latter would suggest “Friday’s rebound merely delayed a repricing to higher real-rate risk.” Nobody quoted in Saturday’s coverage claimed the weekend settled anything, which is the appropriate conclusion to draw from a $445 range.

Macro and cross-asset

There was no Saturday macro session, so the cross-asset marks carried here are Friday’s closes, and their relevance is that they survived the weekend unchallenged rather than that they are new. On TheStreet’s end-of-day figures the ten-year Treasury yield stood at 4.959%, up 1.5 basis points, and the thirty-year at 5.346%, down 1.5 basis points, with TheStreet describing the curve as setting fifty-two-week highs across its length, while Cointelegraph separately reported the thirty-year briefly touching its highest level since June 2004; gold was at $4,404.40, down 0.66%, and WTI crude at $99.05, down 3.35%. Equities closed higher on Friday, though the precise figures require a caveat: TheStreet’s live blog shows midday marks of +1.06% for the S&P 500, +1.15% for the Dow, +0.88% for the Nasdaq and +1.04% for the Russell 2000, where the closing Russell figure carried in yesterday’s brief was +0.63%. The midday and closing marks differ, and this brief flags the discrepancy rather than picking between them. On Saturday itself, CoinGabbar reported US Treasury yields at three-year highs near 4.99%, and ts2.tech put the ten-year’s Friday close at 4.96% while noting that it “keeps the opportunity cost of holding Bitcoin high” — the transmission channel this brief has described all week, in which a higher risk-free yield on dollar cash reduces the relative appeal of an asset that pays none.

The rate expectations are the part of the macro picture that actually had a weekend, and what they did was hold. After August core CPI came in at 0.3% against a 0.2% forecast, CME FedWatch odds of a 25 basis-point increase at Wednesday’s meeting were reported at 85% by Cointelegraph, 85.6% by TheStreet and 86.5% by Decrypt on Friday; Bitcoin News Digest quoted 85.4% for the September 16 decision on Saturday. The cluster is tight enough to treat as one number, and the fact that it neither rose nor faded across a weekend with no new data is itself informative — the repricing from roughly 60% to 69% before the release has settled rather than being second-guessed. The target range is 3.50%–3.75%, so a hike would take it to 3.75%–4.00%. Bitcoin News Digest also reported that EY-Parthenon chief economist Greg Daco had shifted his forecast to project a 25 basis-point increase, which is one named economist moving in the same direction as the futures market rather than an independent confirmation of it. Cross-asset, Ether continued to outperform: CoinGabbar put it at $2,513.18, up 2.6%, against Bitcoin’s 0.6% on the same page’s rolling day, with Ether’s dominance at 11.2% against Bitcoin’s 56.8%. NordFX’s outlook for Ether named $2,515 as the level above which $2,650 opens and $2,300–$2,433 as the zone a hawkish Fed surprise would risk — which places Saturday’s Ether mark within $2 of the level that group identified as the pivot, and makes Wednesday the test for both assets rather than only for Bitcoin.

Calendar: what’s ahead

Date (ET)	Event	Why Bitcoin traders watch it
Sun, Sep 13, 8 p.m.	Weekly candle closes (00:00 UTC Monday)	The frame on which the 50-week averages are defined — the simple average computed here is $79,644.68 and the exponential average named earlier in the week is $77,382.54, which Saturday’s close finished $119.69 beneath. Sunday’s own session is the last input to that close
Mon, Sep 14	US spot ETF flows resume	The first session after a four-day outflow run totalling −$462.7 million that decelerated to −$13.2 million on Friday. There was no Saturday or Sunday session; the next print covers Monday’s trading
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, and a failure could see the bill remade under a different Congress in 2027. The date is carried from multiple published trackers and was re-confirmed on this run
Tue–Wed, Sep 15–16	FOMC meeting; decision Wednesday 2 p.m., press conference 2:30 p.m., with a Summary of Economic Projections	CME FedWatch was pricing a 25 basis-point increase at 85% to 86.5% after Friday’s data and 85.4% on Saturday, against about 60% to 69% before the CPI release; the target range is 3.50%–3.75%. The projections matter as much as the decision, because they carry the committee’s own path for rates into 2027
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, and the last US data the committee sees before it votes
Thu, Sep 17	Bank of England rate decision; US Philadelphia Fed index; final eurozone CPI	The day after the Fed, and the first read on whether other central banks follow the same hawkish repricing that moved Bitcoin this week
Fri, Sep 18	September options expiry; Bank of Japan rate decision; Japanese CPI	The expiry’s notional size was not reported when this brief was written. The BOJ meeting matters for the yen carry trade — borrowing in a low-rate currency to buy higher-yielding assets elsewhere — which has been a live risk to global risk assets all week
~Sat, Sep 19, 12:46 a.m.	Next Bitcoin difficulty adjustment	CoinWarz now estimates +4.25%, to about 132.86T, with 1,019 of 2,016 blocks mined — raised from +3.62% to 132.06T a day earlier. A measured retarget settles the week’s contradictory hashrate readings, because difficulty adjusts on what actually happened over 2,016 blocks
Tue, Sep 29, 10:00 a.m.	August JOLTS job openings	The labour-market side of the case for or against further tightening, two weeks after the decision
Wed, Sep 30, 8:30 a.m.	August PCE inflation and the third estimate of Q2 GDP	The Fed’s preferred inflation measure, arriving two weeks after the decision it will not have informed
Through September	Calendar-month seasonality	September has finished negative in nine of the sixteen completed years in this series and is its weakest month by average return, at −3.3% — though the last three Septembers closed positive. The month is 1.65% lower so far
Ongoing	The long end of the Treasury curve	A thirty-year yield that touched its highest since June 2004 on Friday is the direct competitor to a non-yielding asset. CoinGabbar reported yields at three-year highs near 4.99% on Saturday, so the pressure did not ease over the weekend

Reading the combination

Friday produced the year’s second-largest upper wick; Saturday produced its fourth-narrowest range. Taken together they are a single event and its echo, and the honest reading of the echo is that it carries much less information than its rarity suggests. The temptation with a $445.44 session — 0.577% of the close, 0.189 times the prior day’s ATR, on volume 0.26 times the twenty-day average — is to treat it as a coiled spring, and the base rate appears to support that: of 704 earlier sessions in this series with a range of 0.6% or less while price held above the 200-day average, the close was higher ten sessions later 64% of the time (median +3.08%) against a 58% baseline. But that filter is now substantially selecting for the calendar. All eight of 2026’s sub-1%-range sessions have been Saturdays, and every one in 2024 and 2025 fell on a weekend too — nine Saturdays and three Sundays in 2024, eight and three in 2025; in 2017 the weekend share of such days was 33%, in 2020 it was 55%, in 2023 it was 81%. Weekday volatility has risen until the only sessions quiet enough to clear a low-range screen are the ones when traditional markets are closed. Compared with the right population — 502 earlier Saturdays above the 200-day with a full forward record, which were higher ten sessions later 59% of the time with a median of +1.49% — the 107 quiet Saturdays give 64% and +2.90%. A real tilt remains, about half the size the naive figure implied, on a sample small enough to treat gently. And the narrower question this session actually poses, what follows stillness arriving directly after violence, has seven earlier instances in fourteen years; no percentage should be quoted from seven, and none is quoted here. Beneath that, the layers are where Friday left them, with one addition. The trend layer improved again: a twenty-fifth consecutive close above the 200-day, a 50-day rising for a thirty-second straight session, and a golden-cross gap widening a fifth session from $617.00 to $814.62 — by $197.62, almost four times the day’s own $54.30 price move. The addition is that for the first time in this run the 200-day gained ground on the price on a rising day, climbing $65.97 against the close’s $54.30 and trimming the distance from 10.26% to 10.23%; with $7,169.91 of cushion that is trivia rather than a warning, but it is the mechanical reminder that a rising average eventually reaches a market that stops rising. The momentum layer is still deteriorating and still decelerating sharply: a ninth consecutive MACD widening to −759.80, but by $20.78 after $60.33 and $178.60, two successive reductions of roughly two-thirds, while RSI added 0.17 to 55.05 and the ADX spread gave back Friday’s widening. The flow layer turned from alarming to negligible, −$13.2 million after −$282.7 million, with the four-day run at −$462.7 million and the month still +$307.3 million. The positioning layer quieted in the same way: funding down from 0.0039% to 0.0029%, open interest down $1.64 billion to $51.53 billion, and $104.73 million of liquidations split near evenly instead of Friday’s two irreconcilable readings. Only the macro layer is unchanged and undiminished — CME FedWatch still at roughly 85% for a rate increase on Wednesday after a weekend with no new data to test it, a thirty-year yield at levels last seen in 2004, and the Senate’s CLARITY Act cloture vote the day before the decision. That is the shape of this weekend: five of six layers went quiet, and the one that did not is the one with a scheduled resolution four sessions away. The indicators do not say which way it resolves.

Sources

Price, volume and all technical indicators: Alpha Vantage BTC/USD daily series (00:00 UTC candles), computed by the author. Data note: yesterday’s edition patched the September 11 candle from Coinbase Exchange because the Alpha Vantage feed had served a stale partial row; on this run the Alpha Vantage September 11 row matched that patched candle exactly, to eight decimal places including volume, so the feed is confirmed to have caught up and no patch was needed. The September 12 candle’s completeness is confirmed by the presence of a partial September 13 row. Independent cross-checks against exchange candle endpoints were not possible this run because those hosts were blocked by the network policy in force; the September 12 row was instead validated against the historical distribution of weekend sessions, where its 1,669 BTC of volume and 0.577% range sit within precedent set by prior Saturdays including August 15’s lighter 1,234 BTC and narrower 0.466%. Saturday price marks, the IBIT NAV detail and the $76,850 and $76,030 levels: ts2.tech. Technical survey, oscillator and moving-average counts and the support and resistance bands: Jamie Redman, Bitcoin.com News. ETF flow deceleration, IBIT and per-fund detail, the Bitdeer sale, the 85.4% FedWatch figure and Greg Daco’s forecast shift: Bitcoin News Digest; daily flow history from Farside Investors and SoSoValue. Funding, open interest, liquidations and the resistance zones: CoinStats. Dominance, Ether’s price and share, total market capitalization and the Saturday yield mark: CoinGabbar. Weekly levels, the $69,500–$86,000 base case and the Ether levels: NordFX Analytical Group. September seasonality and the fourth-quarter arithmetic: Motley Fool; the treasury-company argument: Motley Fool. The Revolut data disclosure: Decrypt. Equity, oil, gold closes, Treasury yields, the CPI detail and the 85.6% CME FedWatch figure: TheStreet’s September 11 market blog. Difficulty, block height and the next retarget estimate: CoinWarz. Fear & Greed: alternative.me’s dated API. CLARITY Act vote date: crypto.news. News sentiment tally: Alpha Vantage NEWS_SENTIMENT feed for September 12.