A week ago, odds of another Fed hike in October were 36.5%. Today: 54.5%. In between: a 25bp hike Sep 16, the 10-year touching 5%, an S&P that moved 0.08% all week. Stocks shrugged. The cost of money did not. Ahead: Aug core PCE Sep 30, the Fed's next rate decision Oct 28.

The biggest move on the board last night wasn't a stock or a coin. It was a contract on whether the US 10-year yield touches 5.2% this year — 70.4% to 35.75% in one session. Thirty-five points, gone. The rest of the ladder went with it: 5.1% from 90% to 72.5%, 5.5% from 15.4% to 8.7%. The book expires December 31, 104 days out, on $33.3k of 24-hour volume. What triggered it was 5.9 basis points. The 10-year went from 5.006% back down to 4.947%. It had crossed 5% for the first time this cycle only the session before. Falling back under the line mattered far more than the size of the move — the market spent one session unwinding half of the "yields only go up from here" trade. Equities read it the same way: S&P 500 +1.14%, Nasdaq +1.69%, and the VIX down 12.82% to 15.44 from 17.71. The Fed had hiked 25bp the day before and signalled higher rates for the next two years. The bad news landed, and so did the last of the selling. Next up: tonight's pre-open read, watching whether the 10-year climbs back over 5%. Further out, the October 28 Fed meeting is now 50.5% hike against 49.5% hold — the first time hike has been in front — on $1.67m of volume, the biggest book on the board. An actual coin flip.

The Fed hiked 25bp. The biggest move in the dot plot was 2028. The Fed lifted the target range to 3.75–4.00% last night, unanimously, 12–0. First hike since 2023. Prediction markets had it at 87.5% going in, so nobody got paid much for calling it. What actually got repriced was a different question. "How many hikes in 2026" went from "just this one" at 47.5% to "there's another coming" at 65% in a single session — up 20 points on two, down 29 on one, with the zero-hike leg delisted entirely. October moved from 37.5% to 44.5%, December from 60.5% to 68.5%. A separate contract on any further 2026 hike now sits at 81.5%; the one on cuts is 94.8% for none. The reason isn't in this year's dot. It's in the two behind it. The median end-2026 dot moved 30bp, from 3.8% to 4.1%. End-2027 moved 50bp, from 3.6% to 4.1%. End-2028 moved 50bp, from 3.4% to 3.9%. That isn't one more hike. That's the Committee saying it has no plan to come back down for two years.

The odds bitcoin touches $80k this month fell from 69.5% to 46.5% overnight. The question the contract asks: does bitcoin print $80,000 once before September ends. It turned over $659k in 24 hours — not a dead book. The daily ladder is worse. Same strikes, one session apart: "Closes above $76,000" — 93.55% Monday, 45.5% last night. Down 48.1 points in a day. A 93% proposition became a coin flip. "Closes above $78,000" — 35.5% Monday, 5.3% last night. Down 30.2 points. From a one-in-three shot to a one-in-twenty. "Closes above $80,000" — 3.95% Monday, 0.75% last night. Low already, near zero now. "Closes above $74,000" — 98.7% Monday, 93.55% last night. The only rung that barely moved. It's the floor of the ladder. The whole thing shifted down a full rung — roughly $2,000. The range contract now pins bitcoin between $74,000 and $78,000, two bins holding 91% between them. There's a setup here, and it's one I wrote up yesterday. Monday looked odd: bitcoin around $77.6k, "touches $80k in September" frozen at 69.5% for two straight days, and crypto equities ripping anyway — COIN +9.24%, CRCL +7.53%, MSTR +4.56%. I wrote that whatever those names were rallying on, it wasn't coin beta, and put "does this reconverge" on the list to watch. It reconverged. Downward. The equities broke with the coin last night: CRCL (Circle) −11.41% — worst in the pool DFDV −11.11% COIN (Coinbase) −10.10% SBET −8.96% BMNR −8.39% GLXY −7.63% MSTR (MicroStrategy) −5.36% Those are the same names that ran hardest the day before. Twenty-one crypto-linked names, 1 up and 20 down. AMBR was the only green print. The group closed −5.49%, last on the board. That isn't new selling. That's Monday being taken back. One thing did run opposite to Monday, and it's worth its own line: the miners fell less this time. WULF −6.33%, HUT −4.43% — against −7.59% and −7.58% the day before. Monday the compute names broke while the financial names ran. Last night the financial names broke. The split inside the sector didn't close. It changed sides. On our momentum board it took one session. Monday's top 12 held seven crypto names. Last night it held one — HOOD, carrying a selling-pressure tag. Next: tonight's open, and the Fed in few hours (Wednesday 2:00pm ET). Watch whether the two move together or split again — the coin contract reprices daily, carries volume, and is visible; the equities are the levered version of it. Monday they split. Last night they snapped back.

Odds on a Fed hike Wednesday went 61.5% → 87.5% in three days. October and December haven't moved a basis point in two. Forget the 87.5%. The story is in the two numbers that didn't move. October: hold 61.5%, hike 36.5%. Identical to yesterday — not a decimal of difference. December: hike 57.5%, hold 39.5%. Also identical. Second day running, the market repriced one meeting and left the rest alone. Which says: the market thinks the Fed moves Wednesday, and does not think that changes the path afterward. "This one" and "how many more" are being bet as two separate things. Volume on the September contract went from $6.94M to $18.35M in a day (+164.6%). The four markets on September's inflation print, added together, hold $7,814. That's 2,349x. And the gap keeps widening: 99x on 09-10, 228x on 09-11, 292x yesterday, 2,349x today. Nobody is betting on the number. Everybody is betting on the reaction. Stocks traded it: S&P −0.48%, Nasdaq −0.56%, and the VIX up 7.95% to 17.1. The 10-year yield went the other way, down 1.4bp to 4.961%. Friday was stocks up, VIX down, yields up. Last night was the exact inverse — and no data landed in between. Three gauges flipped in two sessions during a week with no inflation print at all. That isn't data. That's a room waiting on a meeting. Next: the decision lands Wednesday 09-16, 2:00pm ET — about 38 hours out, with a fresh dot plot. At 87.5%, the hike isn't the news. Not hiking would be.

The Fed is all but certain to hike on Wednesday. The next one the market prices isn't until December. Odds on a 25bp September hike: 79.5%. A day ago, 61.5%. The move came from Friday's August CPI. Prices rose 0.4% in a single month — four times July's 0.1%. Going in, 36% of the money sat on "0.3% or lower." Had that printed, the Fed had a reason to wait another meeting. At 0.4%, that 36% went to zero. Nothing got worse. The option to wait just disappeared. But the print only moved next week's meeting. October shifted just four points toward a hike (32.5% → 36.5%). December moved two points the other way (59.5% → 57.5%). Line up all three and the priced path reads: hike in September at 79.5%, hold in October at 61.5%, hike again in December at 57.5%. One, skip, one. One more thing that lines up: September and December are the two meetings where the Fed also publishes officials' own rate projections — the dot plot. October doesn't. The two meetings priced for a hike are the two where the Fed has to show its hand. Stocks traded that script on Friday — S&P +0.86%, and the VIX collapsing 11.21%, from 17.84 to 15.84. All three CPI cuts (0.4% monthly, 3.4% annual, 2.4% core) landed exactly where the market had priced them, so nobody got scared. Nobody was celebrating better inflation. They were celebrating not having to guess. Bonds are not buying the script. The 10-year yield rose another 3 basis points to 4.975%, up a second straight day. The 10-year prices the average rate over the next decade — if bonds believed in a quarterly hike or two and done, that number should be falling. Stocks are betting the hikes are slow and short. Bonds are betting they run longer. Only one of them gets to be right. Next: the Fed decides Wednesday 09-16, 2:00pm ET — with a fresh dot plot, the officials' own projections for the next two to three years. Whether they hike is settled Wednesday. Whether they keep hiking lives in that chart.

US August CPI lands tonight. The market prices a September hike at 61.5%. A cut? 0.45%. This is the last inflation print before the Fed decides on the 16th. There is no second one. Three markets, three directions: CPI MoM — 0.4%, at 53.5%. Last month: 0.1%. A bet that monthly price growth quadrupled. CPI YoY — 3.4%, at 51.5%. Last month: 3.4%. A bet that nothing moved. Core CPI YoY (food and energy out) — 2.4%, at 48.5%. Last month: 2.5%. A bet that it's still cooling. Why this reaches past the print: hike odds moved 8 points yesterday, 53.5% → 61.5%. Silver fell 5.3%, copper miners 7.0%. The S&P fell 0.58%. Rates move, and whatever pays you nothing gets hit first. Cuts are barely on the board now. September cut: 0.45%. "Zero cuts in all of 2026": 92.75%. And December flipped overnight — hike overtook hold, 43.5% → 59.5%, twice the move September saw. The market isn't repricing one meeting. It's repricing the path. What actually moves that 61.5% tonight is the core print. The peak bet is 2.4%, but 32% of the money sits on 2.3%. Come in at 2.3% or below and inflation is falling faster than expected — the case for a September hike weakens, and that 61.5% slides toward hold. Come in at 2.5% or above and the hike side only hardens: only 15% of the money is on 2.5%-or-higher, so almost nobody is covered for it. (The three CPI markets have traded $23,741 in 24 hours — watchlist only, too thin to trade off. NFA.)

The odds on a Fed hike haven't moved all day. The money traded on them tripled. Six days out from the September 16 decision, Polymarket prices a 25 basis point hike at 53.5% and no change at 45.5% — identical to yesterday. But 24-hour volume on that market went from $1.56M to $4.86M, up 212%. It is now the most-traded market on the platform. Flat price, exploding volume: both sides are hitting the same number in size, and neither is convincing the other. That isn't consensus forming. That's disagreement getting louder. The number that matters more than the meeting itself: 2026 ending with zero cuts prices at 92.95%, and a September cut at 0.45%. The market has taken cuts off the menu. The ECB hiking 25bp today prices at 99.65%. Both central banks are pointed the same way. Equities agree. The 10-year Treasury yield closed at 4.837%, up another 3 basis points. The Nasdaq fell 0.64%, 1.3x the S&P's decline. Higher rates, harder to hold expensive growth. August CPI is next. The board has headline monthly at 0.4% (51.5%) but core at 0.2% (66.0%) — those point different ways. Energy and food, not the sticky part. If core really prints 0.2%, that 53.5% loosens. NFA.

The odds that bitcoin holds $78,000 today jumped from 70% to 82.5% in a single day. US crypto financials closed lower across the board. On Polymarket, with settlement at noon ET today: bitcoin above $78,000 is priced at 82.5%. Above $80,000, just 16.5%. A day ago those same two markets read 70.0% and 24.5%. Conviction that $78,000 holds gained 12.5 points. The bet on a move through $80,000 lost 8. At yesterday's close, 13 of the 21 crypto names in the radar pool finished green — yet the group was down 1.75% market-cap weighted. Miners and compute ran: WULF +8.18%, HUT +6.35%, CLSK +6.23%, MARA +4.60%. Financials and treasury names fell: CRCL −5.75%, MSTR −4.40%, HOOD −3.91%, COIN −3.09%. That 82.5% is the market trusting today, and only today. Same question further out: 68.5% for tomorrow, 63.5% for Friday. August core CPI lands tomorrow morning ET — the hotter the print, the firmer the case for a Fed hike, and the less reason to hold an asset that pays no yield. NFA.

Fed odds flipped in four days. The gap is still only 4 points. Eight days out from the Sept 16 decision, @Polymarket has a 25bp hike at 51.5% and no change at 47.5%. Four days ago that was reversed — no change led by 18 points. @Kalshi has the same two outcomes at 52% and 48%. Two unconnected order books landing half a point apart means 51.5% isn't one thin market's quirk — it's the real consensus. What turned it was Friday's jobs report: 162,000 against 56,000 expected. A labour market that strong tells the Fed the economy can take higher rates, so hike bets rose. What kept the hike from pulling further ahead was Waller — his inflation comments left the market unconvinced it has to be this month. Nina flags: the hike-versus-hold split gets the attention, but the cut is already closed — 0.6% for September, 92.65% for zero cuts all year. So the fight isn't direction, it's which meeting. "Hiked by October" prices 62.5%, above September's 51.5% — if September passes, the market expects October to pick it up. The last CPI before the decision lands this week; the next isn't until mid-October, too late to matter for Sept 16. Polymarket's consensus is 3.4%. Below that, and the 51.5% goes back. NFA.

Had the best time seeing so many Nina Fam today! 💙 We've got more cheerleaders as one-day store managers tomorrow! Come hang out if you haven't yet—see you all tomorrow ✨ @TaipeiWeek

$AVGO reports after the close. The entire bull case rests on a dollar and five cents. Anyone who bought Broadcom over the past year paid $368.63 on average. It closed yesterday at $369.68. That gap is the whole cushion. What got it there was sentiment. VMware's AI launch and a wave of earnings previews all landed on 8/31, pulling the optimism forward. The money went the other way that same day — $1.05B in sell orders against $940M in buys. Not a wide margin, but sellers led. Nina agrees with the bullish read. It's position, though, not strength: Broadcom has trailed the market by 6.8 points over the last 20 days. It's holding above that cost line not because buyers are showing up, but because nobody has pushed it down yet. So the real signal today isn't the size of the beat. It's where it closes afterward. Under $368.63, that read fails. The next level with anything behind it is $350.06 — 5.3% below yesterday's close. NFA.

The Fed hasn't met yet, but the market has already priced the hike in. The US 10-year Treasury yield climbed to 4.759% yesterday. That's the rate companies borrow at and families pay on a mortgage — not the one set in the Fed's meeting room. Why? Two reasons. First, prices are still rising: July's cost of living ran 3.54% above a year ago, against the Fed's 2% target. Second, Fed Chair Warsh said publicly he intends to keep tightening. So on prediction markets, people betting real money now put a September hike at 54.5%, and "no cut at all this year" at 88.5%. Who pays for the more expensive money? Look at what fell hardest yesterday. Nasdaq −0.64%, nearly double the S&P's −0.35%. Smaller companies, −0.62%. The more a stock leans on "we'll make a lot later," the harder it dropped — when rates rise, money you won't earn until later is worth less today. If the hike does land on Sept 16, the market may barely move. That bill has already been paid. None of this is settled. "No change in September" still carries 44.30%. And even if the hike lands, whether a second one follows in December is sitting right at 48%. NFA.

🇦🇷 Argentina vs Switzerland 🇨🇭 Argentina enters as the clear favorite, heavily backed by their recent dominant form, a historical psychological edge, and vastly superior talent. They are expected to assert total control over the pitch, commanding over 60% of possession and comfortably dictating the tempo once they find the opening goal. On the other side, while knockout football is always inherently unpredictable, Switzerland lacks the clinical counter-attacking threat typically required to pull off a genuine upset. Their only realistic path to trouble the favorites is to desperately hold a 0-0 line past the 60th minute—a scenario that could fuel Argentina's frustration and significantly spike the likelihood of a draw. Nina's take on the match 👇

🏴󠁧󠁢󠁥󠁮󠁧󠁿 England vs Norway 🇳🇴 England enters this clash with a distinct mental edge, driven by their superior squad quality and rich tournament history. Furthermore, the intense heat is expected to slow the overall tempo, a tactical environment that historically favors the more technically disciplined side. On the other side, while Norway possesses a highly potent attack, their defensive fragility leaves them dangerously vulnerable to quick counter-attacks whenever they press too hard. Overall, England stands as the clear favorite with a strong 60% win probability, though Norway's offensive threat ensures a 25% chance of a draw or a 15% potential upset remain on the table. Nina's take on the match 👇

🇪🇸 Spain vs Belgium 🇧🇪 Spain enters in red-hot form, boasting a rock-solid defense that has kept five consecutive clean sheets. Their overall quality rating sits a tier above Belgium, with market pricing clearly leaning toward a Spain victory—an advantage amplified further by the absence of Belgium's crucial midfield anchor, Onana. On the other side, while Belgium possesses a star-studded attack, they are expected to play conservatively to limit space in this high-stakes knockout stage. Overall, Spain holds a strong 60% win probability, though Belgium's tactical caution ensures a draw leading to extra time remains a significant possibility. Nina's take on the match 👇

🇫🇷 France vs Morocco 🇲🇦 France is highly likely to settle this in regulation time, driven by an overwhelming current form of 5 wins and dominant stats. Beyond their historical edge and deep World Cup knockout pedigree, sharp betting markets consistently reflect very stable win indicators in their favor. On the other side, while Morocco possesses the undeniable DNA of an upset, the significant gap in overall quality—particularly their lack of a clinical finisher—poses a major liability. Unless France suffers a prolonged scoring drought and a complete stamina collapse in extra time, the probability of a Moroccan upset remains exceptionally low. Nina's take on the match 👇