Senate Republicans released what they called the final text of the crypto market structure bill late on Sunday night, with 126 changes that Democrats had asked for. On Tuesday afternoon the Senate voted on whether to begin debating it, and not one Democrat voted yes.

The roll call, from the Senate's own record: cloture on the motion to proceed to H.R. 3633 was rejected 49 to 50, recorded at 2:19 pm, with 60 needed. All 49 yeas were Republicans. Every Democrat present voted no, both independents voted no, and Senator Coons did not vote. Three Republicans voted no on the merits: Collins, Hawley and Moran. A fourth, Tillis, voted no for a different reason. The Senate's Daily Press records that he did it in order to move reconsideration, and he made that motion at 3:01 pm, one minute after the result was announced. A no vote cast to keep a yes vote alive. With every Republican on board the bill would still have been seven short.

The text had been built to win those seven. Senators Lummis, Boozman and Scott said the final version incorporated 126 substantive changes requested by Democrats and substantially all of the Tillis-Gallego ethics language, and AP reported that the President had largely agreed to ethics restrictions covering elected officials, judges and their spouses. Spouses is the word the bill died on. Democrats told The Block that Republican leadership shut down discussions at the last minute, and the point reported as unresolved was whether the restrictions reach past spouses to family members. Lummis, in the release: "Democrats got what they wanted; now they need to take yes for an answer." By Tuesday evening her statement said Senate Democrats had "proved they were never truly serious about protecting consumers and preserving American leadership."

Polymarket's contract on the bill becoming law this year had the answer before the Senate did. It was 34.5% at half past midnight on Monday, on the final text. It held near 30% through Monday afternoon, was 22.5% at five o'clock and 15.5% by seven. It was 13.5% when the roll call began on Tuesday, 4.5% within the hour, and 4.35% late Tuesday. Nineteen points came off on Monday, before a single senator had voted. The vote confirmed it.

Bitcoin's slide started a day before the vote. On CoinGecko's hourly series it peaked at $79,209 at three on Monday afternoon and fell for the next 27 hours, and it was already 2.9% off that high when the roll was called. The vote finished the job: $75,384 at six on Tuesday evening on the hourly series, twenty-eight dollars under Strategy's average cost, with the tick low lower still, and $75,904 late Tuesday, down 2.5% on the day and 4.2% on the week. The stocks priced for a law fell harder. Coinbase, whose chief executive had said publicly he was optimistic it would clear 60 votes, closed up 9.2% on Monday on the final text and down 10.1% on Tuesday, at $172.11. Circle went up 7.5% and then down 11.4%, to $86.30. XRP fell 9.2%, more than three times bitcoin's drop, on the day Ripple's chief executive said "This one stings." Liquidations across crypto ran into the hundreds of millions of dollars over the day, the large majority of them longs, on CoinGlass figures, with futures volume up and open interest down. Those are positions closing.

The honest complication is that the bond market was doing most of the damage before the Senate did any. The ten year Treasury closed Tuesday at 5.00%, its highest close since 2007, after 4.97% on Monday. The two year closed at 4.67%, up 33 basis points since 3 September. The ten year real yield was 2.60% on Monday and 2.62% on Tuesday on Treasury's own table, levels last seen in November 2008. And here is the part almost nobody will print this morning: ten year breakevens went from 2.37% to 2.38%. With Reuters reporting physical crude cargoes in Europe topping $120 a barrel, the bond market raised its inflation expectations by one basis point in a week. Nineteen of the twenty basis points the ten year has added since 8 September were real yield. That is the price bitcoin trades on, and it was moving before, during and after the vote.

And at 2 pm today the Fed decides. Polymarket had a 25 basis point hike at 87.5% late Tuesday, up from 79.5% when we wrote on Monday, with no change at 11.5%. The FILTER is about the part of today that 87.5% does not price.

Three metrics.

Metric 1, Fear and Greed: 51, Neutral. The first reading outside Greed since 19 August, when it printed 46. The stamp taken at eight on Monday evening, after the final text rally, was 69. The stamp taken at eight on Tuesday evening, six hours after the vote, was 51. Eighteen points in a day, and the index went from Greed to Neutral between the roll call and bedtime.

Metric 2, ETF flows: minus $288.6 million on Tuesday, after plus $160.0 million on Monday. Tuesday's print landed late in the evening, and on a day the coin fell 2.5% and Coinbase fell 10% its sign is no surprise. It takes September to about $179 million on SoSoValue's prints. Our 7 October test on the monthly total still tracks Branch A, net positive, but by $179 million instead of the $467 million it was on Monday night, with eleven sessions left to run.

Metric 3, BTC dominance: 58.5%. CoinGecko's global measure late Tuesday, against a total crypto market of about $2.60 trillion, down from the 58.9% we printed Monday even though ether fell further than bitcoin on the day. Strategy's second straight week of no bitcoin and $139.3 million of buybacks of its own preferred is in the RADAR. The one number from that filing that belongs here is $75,412, the company's average cost across 845,050 coins, which bitcoin traded through for an hour on Tuesday evening and sat 0.65% above late Tuesday.

The book is flat for the thirteenth issue running. The STACK retires the CLARITY re-entry rule in writing, and the September hike contract, the one our own rule told us to buy below 40 cents, resolves this afternoon.

📣 THIS SPOT IS OPEN

Baseline Crypto reaches nearly 7,000 Bitcoin-focused readers every Monday and Wednesday morning, opening just under half of what we send. Direct sponsor placements are open, premium position, top of the issue, one per send.

Reply to this email for rates.

The hike is priced. The vote count is not, and bitcoin trades on what the vote count implies.

Take the position seriously, because most of it is right. A 25 basis point increase was 87.5% on Polymarket late Tuesday. The two year Treasury has climbed 33 basis points since 3 September, which is the bond market doing the hike before the committee does. Bitcoin is down 4.2% on the week and gold is also down, which is what assets that pay nothing do when the real yield goes to 2.62%. The June projections already had the year ending at 3.8%, which is one hike from where rates sit, so the move itself has been in the committee's own numbers for three months. And this committee has already shown its hand once: in July it held, with Hammack, Kashkari and Logan each dissenting in favour of a quarter point increase. The hawks lost in July. Today they win. Everyone knows this.

Now look at what 87.5% is a price on. It is a price on the direction of one number at 2 pm. It says nothing about the size of the majority behind it, and nothing about the projections that land at the same moment.

Those two things are what the next year of real yields is made of. Reuters' preview quotes EY-Parthenon's Gregory Daco expecting only one or two dissents in favour of holding, and expecting Governor Waller, who quipped "give disinflation a chance" at a Reuters event before the CPI print, to argue for the hike now. A hike carried over one dissent, with a 2027 median that stays near June's 3.6%, reads as an adjustment: the committee finishing a job it has already priced, and the long end can stay anchored, which is what has happened so far. A hike carried over three or four dissents, or one with a 2027 median that jumps, reads as a campaign against a committee that does not want to run it. The first is bad for bitcoin today. The second is bad for bitcoin for a year, because the real yield in that world does not come back down when the meeting ends.

Run the incentives, because they cut the other way from July. In July the dissenters were hawks who could dissent for free, since a hold with three hawkish dissents costs the chair nothing. Today a dissent is against the chair's first hike, in his first months, with the President on record against it. A governor who dissents today is telling the market the chair does not control his committee, and every trader who prices the path off the dots will read it that way.

What we are not claiming. We do not know the count, and neither does anyone outside the building. We are not arguing with 87.5% on the direction. We are saying the count and the 2027 dot will move the real yield more than the hike will, and that the real yield is the price bitcoin actually trades on. We are also not claiming any edge here: our own published rule said to buy this contract below 40 cents, it traded at 37.5 on 3 September, and we missed it.

So here is the test, graded in Monday's issue. The number is the count of dissenting votes recorded in the FOMC statement released today at 2 pm. Branch A: zero dissents. Branch B: one or two. Branch C: three or more. Three ranges, one published count, nothing outside them. Branches A and B are the world the bond market has been pricing, and B is the one the Reuters preview expects. Branch C is the world it has not priced, and the one in which this FILTER matters.

Three numbers that define the next week.

2 pm today, and the line above the rate line. The statement and the projections land at 2 pm, the press conference at 2:30. A quarter point takes the target range from 3.50 to 3.75% to 3.75 to 4.00%, the first increase since 2023, and Warsh chairs it with Powell casting one of the twelve votes as an ordinary governor. A hike was 87.5% on Polymarket late Tuesday and no change was 11.5%, which pays about eight and a half times the stake if the committee blinks. Read the vote line before the rate line, then read the 2027 dot before the 2026 one. The statement names every dissenter and what they wanted instead, and that sentence, with the 2027 median, is what sets the real yield for the rest of the year. The hike itself is the least informative thing that happens at 2 pm. This is also the moment the September contract resolves, the one our own rule told us to buy below 40 cents. We will grade that on Monday, not today, because today it is still a forecast.

2.65%, credit's answer on CPI day. On Monday we said the single observation that would test our FILTER was how high yield spreads closed on Friday, the day the forecast broke. FRED posted it on Monday: 2.65% for Friday, five basis points tighter on the hot core print, and 2.71% for Monday itself. One observation, and it went our way. Hold the number against 3.00%, which is our entry line and is now 29 basis points away, and against 2.70%, which is the line in our 2 September commitment. At 2.71% on the latest print, Monday's, the spread is one basis point on the right side of that line, and Tuesday's print, which lands mid-morning, comes after a hard day for risk. That is a coin on its edge, and we are calling it a margin for nobody.

About plus 5.1%, the difficulty estimate, now clear of our line. The retarget at block 967,680 is about 460 blocks out and projected for the early hours of Saturday. On Sunday evening the estimate read 3.86%. By Monday midnight it read 4.24%. Late Tuesday, after another run of fast blocks, mempool.space read between plus 5.05% and plus 5.15%, moving with every block. Our own arithmetic off the block timestamps says the remaining blocks would have to average about 596 seconds or slower, roughly the 600 second target, after a fortnight that averaged about 571 seconds, for the final figure to fall back under 4.00%. What to do with it: nothing until Saturday. It grades on the retarget figure on Monday, and the estimate has been on both sides of our line inside four days.

Forget analyst predictions. Polymarket is a real-money prediction market, where traders put actual dollars on outcomes. Scorecard first, then the board. Not financial advice. Our read. Disclosure: we hold personal positions in Polymarket itself and may earn a commission from Polymarket referrals. These markets are thin and the odds below move, so read direction over ticks.

Scorecard: the book is flat for the thirteenth consecutive issue, and one rule is retired today. The only entry condition still standing is high yield spreads through 3.00% into the late October Treasury cash peak, 29 basis points away, and it is covered above.

CLARITY. We bought at 41% and 56% and exited at roughly 33 to 37% on 29 July, a realised loss. The way back in we published had three legs: the merged text survives markup, cloture carries, and the contract is still under 30%. On Monday morning, with the contract at 34.5%, we ruled that the markup leg and the price leg were both unmet, that a cloture win alone would not put us back in, and that if no committee ever marked the text up we would retire the re-entry in print. On Tuesday cloture failed, and Tillis has a motion to reconsider pending, so a second vote is possible this week. That does not revive the rule. The first leg was a committee markup of the merged text. It has not happened, none is scheduled, the House has cancelled its remaining September weeks, and a floor do-over would not count. That is the retirement condition we published. The re-entry rule is retired, in writing, as of this issue. If the Senate passes a market structure bill later, we will write a new rule with a new entry price before the fact, and say so. One thing we will say plainly and without taking credit for it: the rule kept us out at 34.5% on Monday, and that contract was 4.35% by Tuesday night. The legs that kept us out were the markup requirement and the price ceiling, both ruled unmet in print on Monday morning. Neither was a view on the vote, and we had none.

The September hike contract resolves this afternoon. Our published rule said to buy it below 40% with Warsh's position unchanged. It traded at 37.5% on 3 September and we missed it, as we told you Monday. It was 87.5% late Tuesday. If it resolves yes, Monday's ledger records a missed trade that our own rule called. If it resolves no, the rule was wrong and we say that instead. Either way the grade runs Monday, on the result.

Five dated tests are open and a sixth is set in the FILTER. The September ETF monthly total, grading 7 October, is about $179 million after Tuesday's $288.6 million outflow, Branch A but thinly, with eleven sessions left. That figure is on SoSoValue's prints, because Farside, the source the test names, still returns 403 to us; the two matched to within $0.2 million on every day we could compare. The 30 September Treasury cash close, grading 5 October, is in Branch C: the account was $871.2 billion on 14 September, up $53.1 billion from the 10 September low, with Tuesday's corporate tax receipts still to post in this afternoon's statement. The difficulty test grades Monday and sits in Branch A on the estimate, covered above. The high yield spread test, grading 7 October, is in Branch B at 2.71%. The real yield test we set on Monday, grading 7 October on FRED's 30 September print, is in Branch B at 2.60%, with Treasury's own table already at 2.62% for Tuesday. And the 2 September commitment stands: spreads at or inside 2.70% when the Treasury cash peak arrives means our framing was wrong, and at 2.71% we are one basis point from having to say so.

Fed decision in September, 25 bps increase | Market: 87.5% ⚪ NO TRADE, resolves today
No change is 11.5%. Volume across the event is $183.7 million, $37.7 million of it on this leg. Up eight points since Monday, on nothing but the calendar getting closer. We are not buying an 87.5 cent contract with hours to run on an argument, in our own FILTER, that the count matters more than the direction. The count cannot be traded here. The direction can, and it is priced.

Fed rate hike in 2026, any meeting | Market: 95.5% ⚪ NO TRADE, the gap is eight points
September is 87.5%, this is 95.5%. The eight points between them is the whole scenario in which the committee holds today and hikes later in the year. On Monday we printed seven. If the Fed holds and this contract does not fall hard on Thursday, the market is telling you the hike is delayed, and everything priced off the real yield sits through the wait.

Zero Fed rate cuts in 2026 | Market: 93.55% ⚫ EXITED at 88.75%
We shorted this at 78%, stopped out at 88.75% on a published rule, and recorded the loss. At 93.55% on the morning of a hike there is nothing left to debate and nothing to do.

CLARITY Act signed into law in 2026 | Market: 4.35% ⚫ EXITED at roughly 33 to 37%, re-entry rule retired
From 34.5% to 4.35% in under two days, on $20.7 million of volume with $3.4 million of it traded in the last day. A second cloture vote this week would move this, and we are still not buying a four cent contract on a bill that has yet to reach debate. If you want a lottery ticket, buy one where the drawing has a date.

Track everything live at polymarket.com, free, no account required.

Monday's question, whether credit is right or credit is late, drew no replies. We would rather tell you that than invent a consensus, and Friday's 2.65% print answered the first half of it anyway.

Today's question is a number, and it settles at 2 pm.

One question: how many members of the committee vote against whatever the Fed does today? Reply with a number from 0 to 5.

Twelve people vote and a majority carries, so five is the most dissents a decision can have. No reasoning required, though we will read it if you send it. The statement will tell us the answer at 2 pm, and on Monday we will print the spread of your answers next to the real one. If most of you say zero and the statement says three, that is worth knowing about how this list reads the Fed. If most of you say three and the statement says zero, so is that.

Hit reply. We read every response, and the tally runs Monday.

See you Monday with the decision, the dots and the count, Saturday's difficulty retarget graded against our line, and whatever became of Senator Tillis's second vote.

The Baseline Crypto Team

Our three free tools are still down. We said on Monday that a formatting error had broken the wallet check, the custody audit and the public scorecard, and that we would say when they were back. They are not back yet. When they are, it will be in this block, and the scorecard will carry the retired CLARITY rule and the missed September trigger exactly as the STACK does.

In the meantime, this week's is one sentence. Somebody you know keeps their bitcoin on Coinbase and watched the stock fall 10% on Tuesday because a Senate vote went the wrong way. The stock and the custody are two different risks, and the second one is the one they control. Tell them to move whatever they are not actively trading to keys they hold, and to do it on a quiet afternoon, which today is not.