The Weekly BOLT

October 5, 2026

Weekly Recap & Bitcoin Overview

Bitcoin spent the week doing something it hasn't done in a long time: trading on the jobs report. For most of the past two years the first Friday of the month was a stock-market event that Bitcoin watched from the sidelines. This Friday, the September payrolls print came in at 29,000 jobs against roughly 85,000 expected, the unemployment rate ticked up to 4.2%, July was revised to a net loss of 10,000, and within an hour Bitcoin had jumped through $87,000 to a weekly high of $87,249 — the second time in two weeks it has broken $86,000, something it hadn't managed since January. The reason is simple once you see it: the odds of a second Fed hike on October 28 collapsed from about 70% a week ago to around 20% on the print, the 10-year yield dropped 15 basis points in minutes, and every asset that had been squeezed by the September hike got a reprieve. Bitcoin gave back about $2,000 of that spike by midday as the 10-year reversed higher, finished Friday near $84,500, and then climbed through the weekend to about $86,400 by Sunday evening. Call it up a little over 2% on the week from Monday's $84,460 open, with a low of $82,510 on Monday afternoon. The Fear & Greed Index reads 69, still in Greed, down a point from last week.

Monday's low is the other half of the story. Bitcoin dropped to $83,000 after the White House rejected Iran's latest proposal for reopening the Strait of Hormuz, Brent oil climbed back above $100, and gold fell 3.3% in a single session. Roughly $500 million of leveraged positions were liquidated and futures open interest fell to its lowest since March — which, as it turned out, was the setup for Friday. A market with less leverage in it has less to unwind when the news goes the other way, and the holders who bought the $83,000-$86,000 band over the past year, the ones we described two weeks ago as the "wall," got another shot at breakeven and mostly took it again: one widely followed analyst counted about 30,000 BTC sold into the $87,000 area, worth roughly $2.5 billion. That's why the week's high was rejected almost to the dollar of the prior week's $87,363. The band is still the band. But the floor keeps rising underneath it — $82,500 held on Monday where $79,000-$81,000 was the support we'd flagged — and short-term holders are now sitting about 14% above their average cost near $73,700, which is the kind of cushion that makes a cohort patient rather than panicky.

The ETF flows cooled, and we'd rather tell you that plainly than bury it. After nine straight sessions and about $3.1 billion of net inflows — the run that flipped 2026 positive two weeks ago — the spot Bitcoin ETFs saw $149 million leave on Wednesday, mostly from Fidelity's FBTC, and the streak ended. They took in $31 million Monday, $66 million Tuesday, $103 million Thursday, and $32 million Friday, for a net of about $83 million on the week. That's a pause after a $2.4 billion week, not a reversal, and BlackRock's IBIT was positive four of the five days for roughly $290 million. The funds now hold about $109 billion of Bitcoin, roughly 6.4% of all the coins that exist. The more telling number came from Strategy on Monday: 1,665 more Bitcoin bought between September 21 and 27 for $142.7 million, at an average of $85,681, bringing its holdings to 847,666 BTC — paid for with a mix of stock sales and cash, alongside a $152 million buyback of its STRC preferred. The company is adding coins and retiring preferred stock in the same week, at the same price band where the ETF cohort's average entry sits. The treasury bid is steady even when the ETF bid takes a breath.

Now the backdrop, because it's the reason the jobs number mattered so much. The 10-year Treasury yield touched 5.36% on Thursday, its highest since 2007, and the 30-year brushed 5.69%, a level last seen in 2002. Even after Friday's relief the 10-year closed the week at about 5.28%, up 11 basis points, and the 30-year at 5.62%. The 10-year real yield — what Treasuries pay after expected inflation — hit 2.88% on Thursday, its highest since the financial crisis. That number deserves a moment. A June 2023 St. Louis Fed paper by the economist Charles Calomiris warned that, given the size of US debt and projected deficits, if real rates returned to their historical average of "roughly 2%, the US would likely experience an immediate fiscal dominance problem" — meaning the government's borrowing costs start to drive monetary policy rather than the other way around. Real rates are now nearly a point above that line. The Treasury is still paying an average of under 4% across its entire debt stack, so every bond that matures and rolls at 5%-plus raises the interest bill, which raises the deficit, which means more bonds to sell into a market that is already asking for 5.3% to hold them for ten years. That loop is running, and it's why the long end of the curve rose through a Fed hike, through strong PMIs, and through Friday's weak jobs number alike. The inflation data, for what it's worth, came in soft: August core PCE printed 3.0% against 3.3% expected on Wednesday. The bond market barely noticed, because it's pricing supply, not CPI.

Two things from Washington belong next to that. On Wednesday, President Trump told Time magazine that "certain levels of inflation will also pay off that debt very rapidly. Very rapidly," and that the Fed's rate policy is hurting the economy more than inflation is. His National Economic Council director was asked the next day whether the White House plans to inflate its way out of the debt and said "No, absolutely not." We'd simply note that bond investors have long feared exactly the first sentence, and that a 5.3% 10-year yield with a 2.9% real yield is what it looks like when they start pricing it. The second item is quieter but bigger: data this week showed foreign capital flows into US stocks hit an all-time record in the second quarter, while foreign appetite for US bonds keeps fading. East Asia is running a trade surplus near $1.8 trillion a year and, by one prominent capital-flows economist's accounting, almost none of it is being recycled into Treasuries — some goes into US equities, the rest into gold, commodity-linked lending, and spending at home. The US needs that inflow to fund its deficit. When the biggest savers in the world stop buying your bonds and start buying your stocks and your gold instead, you get exactly this picture: equity indices near records, the long bond at 20-year yields, and gold being accumulated by the people who are supposed to be buying your debt.

And then there's oil, which refused to cooperate with anyone. Chinese refiners suspended all October fuel exports on Thursday to protect domestic diesel and gasoline stocks that remain well below pre-war levels, Brent spiked more than $4 on the news, and the G7 and IEA answered Friday with a 100-million-barrel emergency release that knocked WTI back to $91. Brent still finished the week above $102. The Strait of Hormuz remains effectively closed to normal traffic — a fraction of the pre-war 85 transits a day by every tracker, with at least four vessels struck by projectiles this week according to the UK maritime agency — and Iran's parliament speaker said Sunday that it "will not open until our seven conditions based on the Islamabad memorandum are met." Treasury Secretary Bessent's answer last Sunday was that Iran "will have nothing to trade for anything, probably within the next two weeks." Iranian officials, per Bloomberg, are pessimistic about any deal before the November 3 midterms, and the US is sending two more carrier groups to the region. Here is the mechanism we'd put at the center of all of it: as long as oil stays elevated, headline inflation stays elevated, the Fed stays hawkish, real yields stay above 2%, and the debt math gets worse every quarter. Iran appears to understand this perfectly well. A country whose economy is running 90% inflation and whose crude exports have gone to zero is still, by holding the strait, holding the single most important variable in the US bond market.

So what's the through-line? A Fed that hiked into an oil shock it can't fix, a Treasury that can't fund itself at these rates without someone printing the difference, a president who said the quiet part about inflation out loud, and foreign savers who are quietly choosing stocks and gold over bonds. Gold had a bad week in the paper price, down 3.4%, because the dollar hit a two-month high and real yields squeezed it. Bitcoin had a decent one — a higher low, a retest of the highs, and a bid on the first sign the Fed might blink. That's a change in character worth noting without overclaiming: for six weeks now Bitcoin has traded like an asset people buy because of the bond market's problem, not one that gets sold when yields rise. This week it got its first real test of the other direction — yields fell sharply, if briefly, on Friday — and it rallied into that too. An asset that goes up when the Fed is forced to ease and holds when the Fed is forced to hike is doing exactly the job it was built for.

🔦 Market Snapshot (Friday October 2 close / Sunday PM)

Bitcoin: ~$86,400 | Opened the week at $84,460, dropped to a low of $82,510 Monday on the Hormuz rejection and a $500M liquidation flush, then spiked to a weekly high of $87,249 Friday morning on the soft jobs print before settling near $84,500 at the close and recovering to ~$86,400 by Sunday evening. Up roughly 2% on the week. Spot ETFs net +$83M (Mon +$31M, Tue +$66M, Wed −$149M, Thu +$103M, Fri +$32M); the nine-day, $3.1B inflow streak ended Wednesday. ETF holdings ~$109B, about 6.4% of supply. CMC Fear & Greed at 69 (Greed), down from 70. Support: $82,500, then the $79,000-$81,000 zone. Resistance: the $87,200-$87,400 band that's rejected two weeks running. Difficulty was adjusted −0.03% to 132.72T on Saturday — essentially unchanged after September's record increase — with 7-day hashrate near 960 EH/s, down from a bit over 1 ZH/s a week ago, and blocks averaging just over 10 minutes. Next adjustment around October 17-18, currently projected lower.

Gold: ~$4,140/oz spot | −3.4% on the week, a second straight weekly loss, including a 3.3% drop on Monday alone as the dollar rose to a two-month high and 10-year real yields pushed toward 2.9%. About 26% below the January ATH near $5,589. The $4,200-$4,300 floor that held since summer gave way; the next zone is $4,000-$4,100.

Silver: ~$60.36/oz spot | −5.7% on the week, underperforming gold on the way down after leading it on the prior bounce. Gold/silver ratio widened back to about 68.6.

S&P 500: 7,722.72 | +0.7% Friday on the jobs-driven rally but −0.3% on the week; still within 1% of its record. VIX closed at 15.3.

Dow: 51,176.96 | +0.5% Friday, −1.3% on the week, the laggard as rate-sensitive industrials and financials absorbed the Thursday yield spike.

Nasdaq: 27,190.86 | +1.2% Friday and +0.5% on the week to a fresh record, with Nvidia touching an intraday high near $238 and a $5.7 trillion valuation.

Russell 2000: 2,832.89 | +0.9% Friday, −0.2% on the week; still about 8% below its August high.

Oil (WTI): $91.11 | −1.4% on the week after a Thursday spike of $2.45 on China's fuel-export halt was reversed Friday by the G7/IEA 100-million-barrel emergency release.

Oil (Brent): $102.25 | Roughly flat on the week on a like-for-like basis after the contract roll; jumped $4.28 Thursday and held above $100 throughout. Brent-WTI spread near $11, a measure of how much the seaborne market is paying for Hormuz.

10Y UST Yield: ~5.28% | +11bp on the week to its highest weekly close since 2007; touched 5.36% intraday Thursday, then swung from ~5.15% to ~5.30% within hours of Friday's jobs report. Largest quarterly rise in the 10-year since 1994. The 10-year real (TIPS) yield reached 2.88% Thursday, highest since the financial crisis.

2Y UST Yield: ~4.83% | Roughly flat on the week as October hike odds fell from ~70% to ~20% on the jobs print. The 2s-10s spread widened to about 45bp: the selloff is entirely at the long end.

30Y UST Yield: ~5.62% | +13bp on the week; touched 5.69% Thursday, highest since 2002. The MOVE index of Treasury volatility sits above 100, up more than a third in September.

DXY: ~101.9 | +0.7% on the week to its highest since late July, pressuring gold and silver. The dollar rising alongside yields is the combination that keeps FX-hedged Treasury yields negative for Japanese and European buyers — and foreign bond demand away.

💰 Gold, Silver & Hard Assets

Gold fell 3.4% to about $4,140 and silver fell nearly 6% to $60.36, and most of the damage came on Monday, when gold dropped 3.3% in one session as the dollar and real yields jumped together on the Hormuz rejection. That's two losing weeks in a row, the summer floor at $4,200-$4,300 has given way, and gold is now about a quarter below January's record. We'd give you the same honest read we did last week: a 2.9% real yield and a dollar at two-month highs is the worst possible weather for a zero-yield asset, and the paper price is responding to the weather. Investors are selling gold because the Fed is hawkish and real rates are rising. That is a perfectly reasonable thing to do on a three-month view.

Here's why it's probably the wrong thing to do on a three-year view, and it comes straight from the bond market. The reason real rates are above 2% is that the Fed hiked into an oil shock, and the reason that matters is the Calomiris threshold we described above: past about 2% real, the US government's own interest costs begin to dominate everything else, and the historical way out of that is not that rates stay high until the debt is paid down. It's that the central bank eventually buys the debt. Since the Fed began hiking in early 2022 and pushing the country toward that condition, the S&P 500's total return in dollars is up about 76% — and in gold terms it's down about 23%. Gold has outperformed the stock market through the whole tightening cycle, not despite higher rates but because of what higher rates do to a government that owes 120% of GDP. Selling gold on a rate hike in 2026 is the same trade people regretted making in 2022, 2023, 2024, and 2025.

The physical market kept saying the same thing it said last week. The People's Bank of China's most recent monthly addition, 20 tons in August, was its largest since October 2023 and its 22nd straight month of buying; China's gold imports through August exceeded 1,100 tons. The reason that keeps showing up in this section is the capital-flows data we flagged in the recap: East Asia's surplus isn't going into Treasuries, and gold is one of the places it's going instead. Silver's harder week deserves its own note. After leading gold on the bounce two weeks ago, it led on the way down, which is what a metal with a large industrial and leveraged speculative component does when the dollar squeezes. Copper, the other industrial metal we watch, fell 3% on the week even as Chile — about 23% of global supply — reported its lowest monthly output since 2011, down nearly 13% from a year ago on weather and mine disruptions, with strike votes pending at two major mines. A central bank can raise rates to cool demand for copper. It cannot raise rates to make Chilean mines produce more of it. That's the whole problem with fighting a supply-driven inflation with a demand-side tool, and it's why we keep saying the eventual release valve is the currency, not the interest rate.

🌍 Geopolitical Spotlight: Iran Is Holding the Bond Market, and It Knows It

The most important geopolitical fact of the week isn't a battle. It's an interest rate, and the people on the other side of the Strait of Hormuz have figured that out.

Start with what happened. On Monday, the White House rejected Iran's proposal for a phased reopening of the strait, and Bloomberg reported that Iranian officials are pessimistic about any deal before the November 3 midterms — expecting, if anything, escalation afterward. Treasury Secretary Bessent went on Fox News Sunday to say Iran "will have nothing to trade for anything, probably within the next two weeks. They are going to make their final deliveries of oil to China, and then they will have nothing," and that Tehran is "on their knees." The Pentagon confirmed Tuesday that the last US troops have left Iraq after 23 years, a withdrawal agreed under the previous administration and executed by this one, leaving Erbil and the Kurdish north exposed to Iranian strikes by the Kurds' own account; President Trump called Iraq "a very expensive excursion into Hell." Two more carrier strike groups are heading to the region, which would put roughly 20,000 US personnel in theater by late October. On Friday the vice president, secretary of state, defense secretary, and CIA director met at Camp David. On Saturday the Houthis fired missiles and drones at a Saudi oil facility south of Riyadh, and Saudi jets hit Sanaa 26 times in response. And on Sunday Iran's parliament speaker, Mohammad Bagher Ghalibaf, said the strait "will not open until our seven conditions based on the Islamabad memorandum are met," adding that "the era of wasting time and dictating unilateral demands is over." Iran's foreign minister, separately, signaled that inspectors could return to its bombed nuclear sites in exchange for sanctions relief. Through all of it, the strait stayed effectively closed: traffic a small fraction of the 85 daily transits before the war, at least four vessels struck this week, and tankers increasingly running dark.

Now the part that matters for this newsletter. Iran's economy is in ruins — inflation near 90%, the rial down 30%, essentially zero crude loaded onto tankers last month. By any conventional measure Bessent is right that the clock is running out. And yet Tehran keeps choosing to hold the strait, and the reason is visible in Ghalibaf's own social-media feed over the past month, where he has been posting a running commentary on US Treasury yields — congratulating America on a 5.1% 10-year and calling it "the floor two years out," and asking whether a Fed hike "could open SOH or produce a single barrel." Read that for what it is. Iran has concluded that elevated oil prices keep US inflation elevated, which keeps the Fed hawkish, which keeps real yields above 2%, which puts the US Treasury market under the kind of strain we described in the recap. It doesn't need to win the war. It needs to keep oil above $100 until the bond market does the rest. Then China's refiners suspended all October fuel exports on Thursday — ostensibly to rebuild domestic stocks that sit 20 million barrels below pre-war levels for diesel, though it came a week after a Washington summit where fuel was on the agenda — and either way, less diesel on the world market means more upward pressure on prices, and on yields. The G7's 100-million-barrel emergency release on Friday was the West's answer, and it bought a $1.76 drop in WTI. Brent is still above $102.

This is the uncomfortable arithmetic. The Fed raised rates in September to fight an inflation whose single largest driver is a waterway it doesn't control, and in doing so it has effectively handed a piece of US monetary policy to the people who do. Every week the strait stays restricted is a week real yields stay high, interest costs compound, and the deficit grows. The only ways out are a reopened strait, a weaker dollar, or a Fed that decides to backstop the Treasury market despite above-target inflation — and the first of those is in Tehran's hands. That's why the week's two biggest market moves came from a jobs report and a fuel-export notice from Beijing rather than from anything the Fed said. For readers of this newsletter, the lesson is the one we keep returning to: a monetary system whose stability depends on a chokepoint, a trade partner's export policy, and a foreign central bank's willingness to buy bonds is a system with a lot of other people's hands on the controls. The assets that have no issuer, no chokepoint, and no counterparty are the ones that don't care who's holding the strait. This week gold got hit by the dollar and Bitcoin didn't. Over time, both are the same bet.

💹 Rates Recap

The 10-year yield hit 5.36% on Thursday, its highest since 2007, and the 30-year reached 5.69%, a level last seen in 2002 — and then, on Friday's 29,000-job payrolls miss, the 10-year dropped to about 5.15% in minutes before reversing all the way back to 5.30% and closing near 5.28%. Net on the week: 10-year +11 basis points, 30-year +13, 2-year roughly unchanged near 4.83%. The curve steepened again from the long end, and the third quarter closed with the largest quarterly rise in the 10-year yield since 1994. Treasury-market volatility, as measured by the MOVE index, rose more than a third in September and is sitting above 100. Mortgage rates are near 7.3%.

The Fed's week was dovish by its recent standards. New York Fed President Williams said Tuesday that after September's hike "there is no need for urgency," while allowing that "one further upward adjustment of the federal funds target range may be appropriate late this year." Then the data piled on his side: consumer confidence collapsed to 81.9, the lowest reading since 2014; job openings fell to 7.08 million; core PCE came in at 3.0% against 3.3% expected; and the jobs report missed by two-thirds with negative revisions. The one hot number was the ISM manufacturing prices-paid index, which jumped nearly 7 points to 77.9 — its highest since the war began — with 30% of respondents citing Iran by name. Markets took the October hike almost off the table, from roughly 70% to about 20%, and shifted the odds toward December. Atlanta Fed GDPNow has fallen from over 6% in early August to 3.7%.

Here is what we'd ask readers to notice. Growth is slowing, the labor market is softening, consumer confidence is at a 12-year low, inflation prints are coming in below forecast — and the long end of the Treasury curve still rose on the week. That's not a market pricing the business cycle. It's a market pricing supply. The Treasury still pays under 4% on average across its outstanding debt; as that rolls at 5%-plus, net interest is on track to exceed 4% of GDP, more than the defense budget. Veterans' benefits alone have gone from 2% of federal receipts in 2003 to 6% today and are growing 10% a year. Add Social Security, Medicare, Medicaid, and interest together and the mandatory bill already exceeds everything the government takes in. That is the "fiscal dominance" condition Calomiris warned about, and his conclusion was blunt: "at some point people will recognize that at no interest rate will they be willing to accept further government debt issues," at which point "the Fed would be forced to step in and purchase the debt, that is effectively printing money." The question for the October 28 meeting isn't whether the Fed hikes — it probably doesn't now. It's what Chairman Warsh does if the long end keeps rising anyway, which it has done through a hike, a hawkish summer, and a soft jobs report. The answer to that question is the single most important input to every asset in this newsletter.

🤖 AI/Tech Frontier

The Nasdaq made another record and Nvidia touched a $5.7 trillion valuation, so by the index the AI trade is fine. By the credit market it's getting more interesting. A prospectus from one of the leading US AI labs surfaced this week ahead of a planned listing after the midterms, and the document — reportedly 261 pages, roughly 80 of them on risk factors — disclosed about $518 billion of compute commitments, around 80% of them non-cancellable, alongside the lab's own warning that its technology could pose "catastrophic or existential risks to humanity." The same week a rival frontier lab shelved an October model release over safety concerns, saying the model had shown higher rates of deception than its predecessor. We'd note the pattern without editorializing on it: the companies building the technology are now saying in regulatory language what their critics used to say on podcasts, at the same moment they're asking public markets for the largest sums in history.

That's the context for the week's more consequential story, which is debt. The Financial Times reported that AI hyperscalers are now borrowing at such scale, across so many currencies, that other corporations and even sovereign issuers are timing their bond sales to avoid them. KKR warned Wednesday that with tech companies planning nearly $8 trillion of AI infrastructure by 2030, about a fifth of the investment-grade bond index could end up exposed to AI risk — and that the true exposure is larger once off-balance-sheet leases, guarantees, and purchase commitments are counted. "We don't think enough people are talking about the potential volatility if AI growth slows," KKR's co-head of credit told the paper. The Fed chair and the Treasury secretary have both said this year that the hyperscalers are now competing for capital with the $31 trillion Treasury market itself. Put that next to the 10-year at 5.3%: the US government and the largest companies in the world are bidding against each other for the same pool of savings, and the price of that pool is going up for both. One short-seller ran the numbers on a legacy data-center landlord this week and found that refinancing its debt at the current 10-year yield would wipe out its trailing operating income. That's the sharp end of a trend, not the average, but it's the direction.

On the physical side, the buildout keeps finding its real constraint. Amazon signed a 20-year, 690-megawatt power purchase agreement with Constellation's Calvert Cliffs nuclear plant in Maryland on Wednesday, with Constellation committing about $3 billion to upgrades that add another 190 megawatts by the early 2030s — two months after Amazon abandoned a plan for an on-site data center there. Twenty-year nuclear contracts are what you sign when you've concluded the grid won't have power to spare. Chile's copper output falling to a 15-year low is the same story from the supply side: every one of these facilities is wire, transformers, and switchgear, and the metal that goes into them is getting harder to source at the moment demand is accelerating. The frontier labs are competing with the Treasury for capital and with each other for electrons. The bond market is where those two competitions meet, and this week it was the bond market that set the price.

💡 Bitcoin Bite: The Difficulty Ribbon — What Miners Tell You That Price Can't

Saturday's difficulty adjustment was a non-event on its face: a change of three-hundredths of a percent, to 132.72 trillion, after September's record 4.2% increase. But the hashrate behind it tells a different story. The network's seven-day average computing power fell from a bit over 1 zettahash a week ago to about 960 exahashes this week, and the next adjustment in mid-October is currently projected to be lower. Hashrate fell while price held. That combination is one of the most useful signals in on-chain analysis, and the tool built to read it is this week's topic: the Difficulty Ribbon.

The Basics: Bitcoin's difficulty is the number that determines how hard it is to find a valid block, and it resets every 2,016 blocks — about two weeks — so that blocks keep arriving roughly every ten minutes no matter how much computing power is pointed at the network. More miners, higher difficulty; fewer miners, lower. That makes difficulty a slow, honest record of how much real-world capital and electricity is being committed to the network over time. The Difficulty Ribbon, developed by analyst Willy Woo, takes that series and plots several moving averages of it at once — typically the 9-day, 14-day, 25-day, 40-day, 60-day, 90-day, 128-day, and 200-day — on the same chart. When difficulty is rising steadily, the averages fan out: the short ones run ahead of the long ones and the ribbon is wide. When difficulty stalls or falls, the short averages drop back toward the long ones and the ribbon compresses, sometimes inverting entirely. The width of the ribbon is the whole signal. Wide means miners are expanding; compressed means they're under stress.

Why It Matters: Miners are the only participants in Bitcoin who are structurally forced to sell. They have electricity bills in dollars and revenue in a coin whose price they don't control, so in the normal course of business they sell a portion of what they mine. When the price falls or difficulty rises faster than price, the least efficient miners stop making money, switch off their machines, and — critically — sell whatever Bitcoin they've been holding to cover costs on the way out. That capitulation is what a compressed ribbon captures: hashrate dropping, difficulty following, weak miners leaving. The reason this has historically been one of the better long-term buy signals is what happens next. Once the weak miners are gone, their forced selling is gone with them. The miners that remain are the efficient ones with cheap power, who can afford to hold rather than dump. Supply pressure from the one cohort that has to sell drops sharply, and the price tends to find a floor. Woo's original observation was that ribbon compressions have marked most of the major cycle bottoms, and that the periods just after the ribbon re-expands have been among the best stretches to own Bitcoin. The signal isn't about miners being smart. It's about the market losing its most reliable seller.

This Week's Connection: The ribbon isn't compressed yet — difficulty is still at a record and only went flat — but this week was the first hint of the short averages starting to roll over. The hashrate decline from around 1.03 ZH/s to 960 EH/s is roughly 7%, and it's happening with the price near $86,000, not in a crash. There are two possible reads. The benign one is seasonal and operational: the end of the quarter, maintenance, power curtailments in Texas and elsewhere as the grid tightens, miners selling capacity back to the grid during the demand peaks we keep writing about in the AI section. The more structural one is margin: miner revenue per unit of hash sits near its lows after September's 4.2% difficulty jump, and at $86,000 with the record difficulty, some older-generation machines are now marginal. Watch the mid-October adjustment. If it comes in meaningfully negative — the projection is a few percent lower — and hashrate keeps sliding into a flat price, the ribbon starts to compress and the older hardware is leaving. That's a temporary headwind in miner selling, and then, historically, a tailwind once it's done. What makes this cycle different is who the marginal buyer of that selling is: ETFs that took in $3.1 billion over nine days, and a corporate treasury that bought 1,665 coins last week at the same price.

The Long-Term Lesson: The Difficulty Ribbon is a reminder that Bitcoin's price has a physical layer underneath it. Price is what traders agree on; hashrate is what miners have actually paid for in electricity and machines. The two can diverge for a while, but miners are the participants with the least ability to pretend, because their bills come due every month. When the ribbon compresses, the network is shedding its weakest hands at the cost basis where it hurts most, and the holders who understand that have historically been the ones buying from them. For the patient reader, the practical read is the same one it's been in every Bite this year: the cohorts that are forced to sell — leveraged longs, breakeven holders, marginal miners — are the ones that create the opportunities, and the cohorts that don't have to sell are the ones that end up owning the supply. This week the ribbon flickered. It hasn't spoken yet.

🚨 What to Watch in the Weeks Ahead

The $87,300 ceiling and the $82,500 floor: Two weekly highs within $120 of each other and a Monday low that held exactly where last week's support was. The band is well-defined. Watch ETF flows: a return to daily inflows above $100 million would give the third test of $87,000 the fuel the first two lacked. Renewed outflows and a loss of $82,500 would open the $79,000-$81,000 zone.

ISM Services (Monday, Oct 5): Manufacturing's prices-paid index just hit a war high. If services prices follow, the soft-PCE story has a problem, and so does the dovish repricing from Friday.

September CPI (Wednesday, Oct 14): The last major inflation print before the October 28 FOMC. With a hike mostly priced out, a hot number would reverse Friday's relief quickly — and with the 10-year at 5.3%, the long end doesn't need an excuse.

The mid-October difficulty adjustment (~Oct 17-18): Currently projected lower. A meaningful decline with hashrate still sliding is the Difficulty Ribbon starting to compress — short-term miner selling, longer-term floor-building. See this week's Bite.

Hormuz, tankers not talk: Iran has restated its seven conditions; the US has sent two more carrier groups; Iranian officials don't expect a deal before November 3. Transit counts are the only number that matters for oil, and oil is the only number that matters for real yields.

October 28 FOMC: The hike is largely priced out after Friday. The question is the same as last week, with the stakes higher: does the long end rise on a hold the way it rose on a hike? If it does, the bond market is telling the Fed that the problem was never the funds rate.

China fuel exports and the G7 release: Beijing's October export suspension is the first of its kind since March. Whether it's extended into November, and whether the G7's 100 million barrels arrive in time to matter, will set the diesel price into winter.

Foreign flows: Record foreign buying of US stocks and fading foreign demand for US bonds is the structural story under the whole curve. The next TIC data release will show whether the September yield spike drew any of that money back into Treasuries — or whether it kept going into gold.

📅 Upcoming Events

Bitcoin Amsterdam 2026: November 5–6 | Sugar Factory, Amsterdam
Two days after the US midterms, with 30-year Treasury yields at levels last seen in 2002 and the question of who funds Western deficits now a live one in Europe too.

APEC Leaders' Summit: November | Shenzhen, China
Where the US-China AI dialogue holds its first talks and where Beijing's fuel-export policy, now a tool of leverage, will be on the table alongside trade.

Bitcoin Abu Dhabi 2026: December 2026 | Abu Dhabi
At the crossroads of the capital flows that define this decade, with the Hormuz question still the single most important variable for oil, inflation, and the bond market.

✅ Closing Thought

This was a week where the jobs report mattered to Bitcoin, which is itself a sign of how the asset has changed. A 29,000-payroll miss took a Fed hike off the table, knocked 15 basis points off the 10-year for an hour, and Bitcoin spiked to $87,249 on it — then gave the spike back when yields reversed, held $84,500, and climbed back to $86,400 over the weekend. Two weeks ago it broke $86,000 on the largest ETF inflow day of the year; this week it did it on a macro print, with the ETF bid resting. The wall of breakeven holders at $87,000 is still there and sold into it again. But the floor moved up to $82,500, short-term holders are sitting on a 14% cushion, the leverage got flushed on Monday, and Strategy added 1,665 coins at the same price everyone else is arguing about. That's a market consolidating from strength, not weakness.

The macro, meanwhile, got clearer and worse at the same time. The 10-year real yield touched 2.9%, past the threshold a St. Louis Fed paper warned would put the US in "immediate fiscal dominance." The Treasury is still paying under 4% on debt it now has to roll above 5%. The president said inflation would pay off the debt "very rapidly," and the bond market, which has feared exactly that for years, took him at his word. Foreign savers are buying American stocks and gold instead of American bonds. Iran is holding the strait, apparently because it understands that oil above $100 does more damage to the US through the bond market than it ever could through the oil market. And China turned off fuel exports in the middle of all of it. The Fed hiked into this, and the long end rose anyway. Soft jobs, soft inflation, collapsing confidence — and the long end rose anyway.

Gold caught the dollar on the chin this week and gave back 3%. We'd say what we said last week: that's the weather, and the climate is a 2.9% real yield on a government that cannot afford one. Bitcoin, meanwhile, went up on the day the Fed was forced to blink and held on the days it wasn't. We'd keep watching the Difficulty Ribbon, the $87,000 band, and the long end of the curve on October 28 — because whatever Chairman Warsh decides about the funds rate, the market has already told us what it thinks about the debt. We'll be back next Monday.

Pass it on. If this issue saved you an hour of headlines, forward it to someone who'd get something out of it. Three referrals through your link unlocks The Bitcoin Bites Collection — every on-chain tool we've written up, in one PDF. Ten gets you a shout-out in the newsletter. New here? Our free guide Bitcoin, Explained — ten short chapters on what Bitcoin is and why it exists — is at staybolted.com/bitcoin-explained.pdf. And if you'd rather say thanks in sats, there's a Lightning address at staybolted.com/tip.

As always,

Stay Bolted — BOLT Team

staybolted.com

Disclaimer: The Weekly BOLT is published by Stay Bolted, LLC and is provided for informational and educational purposes only. This newsletter does not constitute financial, investment, legal, tax, or any other form of professional advice. All content, including discussions of Bitcoin, macroeconomic trends, and related topics, is based on publicly available information and the opinions of the authors, which may change without notice. We make no representations or warranties as to the accuracy, completeness, or timeliness of the information presented. Readers should conduct their own research and consult with qualified professionals before making any financial decisions. Investments in Bitcoin and other cryptocurrencies involve significant risks, including the potential for total loss of principal, and are not suitable for everyone. Stay Bolted, LLC, its affiliates, and contributors disclaim any liability for losses or damages arising from reliance on this newsletter. Past performance is not indicative of future results.