Oil tops $107, the October hike reaches 66%, and Bitcoin stalls below $87,500

Oil tops $107, the October hike reaches 66%, and Bitcoin stalls below $87,500

Oil back above $107 after Washington rejected Iran's Hormuz offer repriced the October Fed meeting to 66% for a hike, knocked Bitcoin's $87,500 rung down 60 points on the same yield move, and left Brazil trading 16 points away from its own polls before Sunday's first round.

Oil did the work this week. Crude went back above $107 a barrel after Washington turned down Tehran's offer to reopen the Strait of Hormuz, and that single price repriced everything downstream of it. The October Federal Reserve meeting went from a coin flip to a two-to-one favourite for a rate hike, the 30-year Treasury yield touched its highest level since June 2004, and Bitcoin's push at $87,500 and $90,000 collapsed while spot itself gave back only a few percent. Oil was the week's single thread, running through the Fed, the Strait of Hormuz and crypto. Brazil and two American races moved on their own news.
This recap covers September 21, 9:00 a.m. through September 28, 2026, 9:15 a.m. ET. Polymarket prices, weekly changes, volume, liquidity and order-book quotes were refreshed between 9:07 and 9:15 a.m. ET on September 28. An inferred starting probability equals the current Yes price minus Polymarket's displayed one-week change, so it is a derived figure rather than a recorded one, and dollar amounts are rounded to the nearest dollar. Polymarket's public market records stop at probability, volume, liquidity, best bid, best ask and spread. Authenticated Yes-versus-No net flow and a dependable large-trade list sit outside what Polymarket publishes publicly, so the market-structure checks below use book depth and the quoted spread, and this edition makes no claim about which side of a trade the money came from.

The board at a glance

Theme and contractCurrent YesInferred Yes one week agoVolume in past 7 daysLiquidityBest bid / askResolution clock
Fed: +25 bp after the October meeting65.5%49.5% (+16.0pp)$1,798,956$389,81165¢ / 66¢October 27–28 meeting 12
Fed: no change after the October meeting32.5%48.5% (-16.0pp)$1,727,047$323,19432¢ / 33¢October 27–28 meeting 3
U.S.-Iran ceasefire continues through September 3096.8%84.5% (+12.3pp)$1,588,226$171,53896.7¢ / 96.9¢September 30 4
U.S. announces an end to the Iranian blockade by December 3158.75%63.15% (-4.4pp)$509,254$188,28758.5¢ / 59¢December 31 5
Strait of Hormuz traffic returns to normal by December 3123.5%19.5% (+4.0pp)$1,395,171$260,98823¢ / 24¢December 31 6
Bitcoin reaches $87,500 in September10.5%70.5% (-60.0pp)$728,927$80,48110¢ / 11¢September 30 7
Bitcoin dips to $80,000 in September15.15%29.6% (-14.45pp)$403,264$83,17414.7¢ / 15.6¢September 30 8
Flávio Bolsonaro wins Brazil's presidential election57.65%60.2% (-2.55pp)$832,218$370,07557.6¢ / 57.7¢October 4 9
Nithya Raman wins the Los Angeles mayoralty58.45%37.3% (+21.15pp)$492,670$154,81958.4¢ / 58.5¢November 3 10
Four of those rows sit inside bigger boards, and the two kinds of number measure different things. The Fed contract trades inside "Fed Decision in October?", which turned over $6,602,680 in the week on $2,161,795 of book liquidity, and the December board ran $595,589 with $317,669 of open interest. 1112 The two blockade contracts share an event with $2,887,758 of weekly volume and $2,374,579 of open interest, the ceasefire contracts a board with $2,278,992 and $823,023, the September Bitcoin ladder a board with $5,315,402 and $3,437,626, and Brazil's presidential election a board with $3,395,342 and $7,179,967. 13141516 A child row counts trades in the contract named in that row; a board counts every threshold in the series. The two are kept apart for that reason.

The Fed's October meeting turns into a two-to-one favourite for a hike

The FOMC lifted its target range by a quarter point to 3.75%-4% earlier this month and said more increases were likely. 17 Polymarket's contract on a further quarter-point increase after the October 27-28 meeting moved from an inferred 49.5% to 65.5% in seven days, up 16.0 points on $1,798,956 of weekly volume, with a 65¢ / 66¢ quote inside a $389,811 book. 12 The mirror contract fell 16.0 points to 32.5% on $1,727,047. 3
Rate futures moved with it. CME's FedWatch put the implied probability of an October increase near 66% late in the week, against roughly 58% seven days earlier, and the implied probability of one increase by December at 93% to 94%. 1819
The catalyst chain ran through the bond market before it reached the prediction market. On September 23, S&P Global's flash U.S. business survey put output growth at its fastest pace in more than five years, with a composite reading of 58.4, and the Treasury's $70 billion five-year note sale cleared at 5.033%, the highest auction yield since 2006 and about three basis points above where the notes traded beforehand. 20 By Thursday the 30-year yield had touched 5.5016% intraday, its highest since June 2004, and closed at 5.472%, while the 10-year reached 5.2251%, a 19-year high, and the two-year sat at 4.914%. 18 Cleveland Fed President Beth Hammack said on Friday that she worries persistently high inflation will condition the public to treat elevated prices as normal, and that the central bank cannot allow that to happen. 17 Philadelphia Fed President Anna Paulson, New York Fed President John Williams and Governor Michael Barr also said further increases may be needed. 18
Then oil arrived. After the Hormuz rejection, Brent traded back above $107 on Monday morning and spot gold fell 1.5% to $4,223.95 an ounce, with the dollar near a two-month high. 1721 Higher crude raises costs across the economy, which pushes the inflation path up and the rate path with it. 17
The December board agrees with the direction and disagrees about the size. A quarter-point increase there trades at 78%, up 10.5 points, while no change fell 10.0 points to 19.5% and a half-point increase sits at 2.05%. 212
One signal from this week is worth reading with caution. Polymarket's contract on the 30-year Treasury yield reaching 5.50% in September jumped 88.45 points to 91.45%, and it carries only $19,009 of weekly volume, $242 of book liquidity and an 11.9-cent spread between its best bid and best ask. 22 The underlying fact is well documented; the contract itself is too thin to trade and too wide to read as a crowd signal.
Hike case (65.5%): the Fed itself flagged more increases, three more officials said the same this week, the 30-year cleared 5.50% intraday on Thursday, and crude above $107 makes the inflation problem harder rather than easier. 1821
Hold case (32.5%): the October meeting lands six weeks after a hike whose effect has yet to show up in any data, and this week brings the September PCE index and the payroll report, the two prints most likely to move the October price. 1718
Conditional trade inference: at 65.5¢ the October contract already pays the hike, so the information sits in the timing curve. If October happens, December at 78% makes a year-end funds rate of 4.25%-4.50% the base case, and that is what the front end, the dollar and gold have spent the week repricing. The branch that would kill it arrives in the next few sessions: a soft September PCE reading or a weak payroll number pushes the October contract back toward even money, and long-duration Treasuries, gold and rate-sensitive equities are where that relief would show up first. 1217

Iran's seven-day offer gets rejected, and the ceasefire holds while the blockade stays

Negotiators for Washington and Tehran met in New York during the United Nations General Assembly and worked on a phased path out of a conflict now nearly seven months old. The most plausible arrangement under discussion was staged: the United States lifts its economic blockade, and Iran reopens the Strait of Hormuz, with Tehran potentially gaining access to frozen assets. Regional officials said Iran was willing to move its demand for transit fees into a side attachment because ending the blockade matters more to it, while Gulf leaders rejected any Iranian control of the waterway and pressed for freedom of navigation. 23
Iran's foreign minister, Abbas Araqchi, described the proposal as a seven-day countdown to reopening the strait plus a pause in regional fighting, with wider talks on "mutually agreed subjects" to follow. 24 Traders read the week that way at first. Oil fell about 2% on Friday, with Brent settling at $104.32 and WTI at $92.41, on truce hopes and on talk of a possible U.S. ban on diesel exports. For the week Brent ended up less than 1% while WTI lost about 8%, and the premium of Brent over WTI reached its highest since May. 25
Trump rejected the proposal on Saturday. He said Iran needs a deal because it is losing badly. His envoy Mike Waltz said the offer asked for immediate lifting of sanctions and release of frozen assets, and that Iran wanted everything up front in exchange for a promise to talk later. Araqchi responded that mediators had not officially relayed any rejection and that only a negotiated solution ends the deadlock, while President Masoud Pezeshkian said Iran is ready to discuss its nuclear programme but will accept no coercion. Trump told Axios on Sunday that he expects more talks this week. 24 Oil went back up about 3% on Monday, with Brent at $107.30 and WTI at $95.22 by 1206 GMT, and greater flows through the strait took some of the pressure off while leaving the market in deficit. 2126
The prediction markets split the two questions apart. The September ceasefire contract rose 12.3 points to 96.8% on $1,588,226 of weekly volume, because it settles on one narrow fact: whether the United States carries out an air strike or a surface-to-surface missile strike that directly impacts Iran before the deadline. Ground incursions, naval gunfire, artillery and cyber operations fall outside the rule. 4 The same question pushed out to October 31 trades at 56%, up 3.5 points, which is a large discount for four extra weeks of no strikes. 14
The blockade contracts answer the other half, and this is where the rejection shows up. A U.S. announcement ending the blockade by September 30 fell 4.5 points to 5%, the October 31 contract fell 8.0 points to 34.5%, and the December 31 contract fell 4.4 points to 58.75%, all on the rule that a qualifying statement must express a general cessation of the blockade rather than an exemption for one vessel, cargo or port. 52728
Two more rows on that board are worth keeping in view. The September contract on Strait of Hormuz traffic returning to normal has all but expired at 0.25%, yet it was the largest non-sport turnover on the platform this week at $2,328,299, which tells you how much money was positioned for a physical reopening; it needs IMF PortWatch to publish a seven-day moving average of transit calls at 60 or above. 29 The December version of the same question, built on the same PortWatch threshold, rose 4.0 points to 23.5% on $1,395,171 — so the board gives even a four-month reopening a little better than one chance in five. 6 The Saudi East-West pipeline restart contract sits at 31.5% on $477,975 of weekly volume with a $36,167 book and a three-cent spread; Polymarket publishes no seven-day change for it. 30 "Will the U.S. invade Iran before 2027?" moved a single point to 15.5% on $2,227,169 of weekly volume and $752,754 of book liquidity. 31
Crude flows out of the strait reached 33.7 million barrels in the week starting September 20, and Houthi attacks on Saudi Arabia continued alongside the diplomacy. 25
Ceasefire case (96.8%): two rounds of talks happened in New York, the president said he expects them to continue this week, and the contract asks about one specific class of American strike rather than about the conflict as a whole. 24
Ceasefire, other side (3.2%): the two boards disagree about October, where the same question prices at 56%, and Houthi strikes on Saudi facilities keep the escalation path open while American naval pressure continues. 1425
Blockade case (58.75% by December 31): both sides spent the week describing the same sequence — blockade relief first, strait second — and Trump named a window after the November 3 midterms, when lower gasoline prices would help him politically. 23
Blockade, other side (41.25%): Washington turned the staged offer down on Saturday and said Iran wanted sanctions relief and frozen assets before any concessions, and the September and October contracts on the same question fell 4.5 and 8.0 points. 2428
Conditional trade inference: the ceasefire row answers whether Washington strikes, and the blockade row answers whether Washington announces; crude is where both questions land. A framework before the midterms points at the mid-$90s for Brent, which favours transport, airlines and consumer discretionary over energy producers. A stalemate past the September 30 expiries puts the high rungs of the oil ladder and the December Hormuz-traffic contract back in play, and the energy sector plus inflation-linked exposure is where that branch pays. The date that decides between them is the one Trump named himself. 621

Bitcoin's run stops at $87,500, and the ETF money keeps coming anyway

Bitcoin opened the week strong and spent the rest of it giving the gain back. Kraken's daily BTC/USD candle for September 21, which opens at 7:00 p.m. ET the previous evening, printed a high of $87,446.70 and closed at $86,593.80. By September 23 it closed at $84,384.60, and on the morning of September 28 it traded at $83,514.80. 32 The September ladder settles on any Binance BTC/USDT one-minute candle high at or above the strike between 00:00 ET on September 1 and 23:59 ET on September 30, so venues can print slightly different extremes. 7
That single dollar gap mattered. The $87,500 contract fell 60.0 points to 10.5% on $728,927 of weekly volume with an $80,481 book, the $90,000 contract fell 39.4 points to 2.5% on $606,430, the $92,500 contract fell 23.8 points to 0.95%, and the $95,000 contract fell 12.95 points to 0.5%. On the downside, the $80,000 dip contract rose to 15.15%, the $77,500 contract sits at 4.1% and the $75,000 contract at 0.9%. The whole September board turned over $5,315,402 and carries $3,437,626 of open interest. 81533
The cause was in the bond market rather than in crypto. The September 23 business survey and the weak five-year auction pushed the 10-year yield up 15 basis points to close at 5.11%, and Bitcoin's steepest drop that day came shortly after the survey landed; the following morning it traded near $83,900, down more than 2% over 24 hours after touching close to $87,300. Dogecoin led a broad sell-off with a 7% fall, and ether, solana and BNB each lost 2% to 3%. A roughly $14 billion Deribit options expiry fell on September 25, with a large block of calls struck at $85,000. 20 Higher government yields raise the cost of holding an asset that pays nothing, and of borrowing against a leveraged position in it. 20
The flows run the other way. U.S. spot Bitcoin funds took in money on all five sessions from September 21 through September 25 — +$999.0m, +$714.7m, +$346.9m, +$190.7m and +$134.5m — for a net +$2,385.8 million across the week. The money arrived steadily and shrank each day while spot fell from $86,593.80 to $84,090.50 over the same five sessions. 3234 Both facts come from their own sources and are reported side by side; the daily fund data shows what was bought, and the price data shows what spot did with it.
Ether followed the same shape with less amplitude, closing at $2,775.84 on September 21 and $2,687.40 on September 28. 35 The contract on ether dipping to $2,250 by December 31 sits at 30.5%, down 3.5 points, on $545,366 of weekly volume. 36
Yes case on $87,500 (10.5%): ETF money arrived every day of the week for a net $2.39 billion, and the contract needs one Binance minute anywhere in the next two days, which is a weaker test than a close. 734
No case on $87,500 (89.5%): the high that came closest was set on September 21, spot has since given back $3,079 from that close, and the session after the business survey began the slide while the 30-year and 10-year yields stood at multi-decade highs. 1832
Conditional trade inference: the ladder pays on a print rather than a trend, and the top rungs now pay almost nothing, so the live question is which direction is still open. The $80,000 dip contract at 15.15% needs another 4.2% down in two days; the $87,500 rung needs 4.8% up. The PCE reading and the payroll report land inside that window, and they move the real yields that have been pulling crypto around all month — a soft pair revives the upper rungs, a hot pair makes $82,500 at 68.5%, the level the market already treats as more likely than not. 817

Brazil goes to the polls on Sunday with the board and the pollsters 16 points apart

The Brazilian first round is on October 4, with a runoff on October 25 if nobody takes more than half of the valid votes. 37 Polymarket prices a Flávio Bolsonaro win at 57.65%, down 2.55 points on $832,218 of weekly volume with a 57.6¢ / 57.7¢ quote, and a Lula win at 41.5%, up 2.0 points on $942,423. Both contracts resolve on the winner including any second round, on a consensus of credible reporting or the official count from Brazil's Superior Electoral Court. 938
Three polls landed inside the week and all three describe a tied race. Datafolha, with 2,002 interviews on September 22-23, put Lula at 47% and Flávio at 45% in a runoff, inside its two-point margin of error, and Lula ahead 40% to 36% in the first round. 39 AtlasIntel's poll for Bloomberg, with 5,015 interviews on September 17-22 and a one-point margin, had Lula at 47.7% and Flávio at 47.4% in a runoff and Lula ahead 45.8% to 43.4% in the first round, reversing the same pollster's September 17 reading that had Flávio ahead 47.2% to 46.8%. 37 Genial/Quaest, with 2,004 interviews on September 24-27, put the runoff at 41% each, having moved Lula up a point and Flávio down a point, and the first round at Lula 37%, Flávio 33%. 40
A tie in the polls against a 57.65% price is the widest poll-versus-market disagreement on this week's board. 940
Flávio case (57.65%): every published survey keeps the first-round leaders short of 50%, so the contest runs long, and a campaign that extends to October 25 gives the challenger another three weeks to consolidate the conservative field and the anti-incumbent vote. 40
Lula case (41.5%): the two polls with the largest samples both put the incumbent ahead in the first round and ahead or level in a runoff, Datafolha showed no movement toward the challenger across a week, and Quaest — the institute that had Flávio ahead in a runoff earlier in the month — closed that gap to zero. 373940
Conditional trade inference: the 57.65% price answers two questions at once — who wins, and whether the campaign runs past October 4. A reader who trusts the polls is looking at a first round that settles nothing and a second round priced near even money; a reader who trusts the board is looking at a challenger who wins outright. Whoever they side with, the October 4 result turns a 16-point disagreement into a settled fact, and Brazilian real-denominated rates and the local equity index both carry more of that event risk into the weekend than a tied campaign would normally justify. 916

Two American races reprice the week the polling landed

Los Angeles moved hardest of any non-crypto market on this week's board. Nithya Raman rose 21.15 points to 58.45% on $492,670 of weekly volume, while Mayor Karen Bass fell 21.0 points to 41.5% on $291,107, inside an event that turned over $783,777 with $293,907 of book liquidity. The contract resolves on whoever wins, and with mail ballots going out on October 1 that decision lands on November 3. 104142
A Berkeley IGS/Los Angeles Times poll released on September 23, with 2,100 likely voters, put Raman ahead 39% to Bass's 28% with about a third of voters undecided, and found Bass at 63% unfavourable against 28% favourable, while Raman sat at 40% favourable and 39% unfavourable. The poll director attributed much of the undecided bloc to primary voters who backed Spencer Pratt, who took more than 217,000 votes in June, and Bass's summer included a rising homelessness count and the Lineage fire in Boyle Heights. 43
Maine moved the other way, toward the Republican incumbent. The contract on Democrats winning the Senate seat there fell 10.0 points to 59.5% on $284,014 of weekly volume, while Republicans rose 10.0 points to 40.5% on $281,770, inside an event with $565,784 of weekly turnover and $266,927 of book liquidity. 444546
Two polls published within a day of each other point in opposite directions. The New York Times/Siena survey with the Portland Press Herald, 619 respondents polled September 15-22 with a margin near five points, put Senator Susan Collins ahead of Troy Jackson 49% to 46%. A UMass Lowell/YouGov poll of 650 likely voters, issued the same day on fieldwork from August 28 to September 14, put Jackson ahead 48% to 43% within a 4.66-point margin. Decision Desk HQ had Jackson at 54% as of September 23. 4748
Raman case (58.45%): the freshest poll has her 11 points ahead with the ballot going out in days, the incumbent's unfavourable rating runs more than two to one against her favourable one, and this month is the first time Raman has led Bass in a poll. 43
Bass case (41.5%): about a third of the electorate is undecided a month before the vote, and 58.45% is a long way above the 39% share the poll gave Raman. 43
Jackson case (59.5%): the UMass poll has him ahead, the race has been inside the margin all year, and a state that has elected a Republican senator for six terms is being priced at 40.5% by the board even as some public models put Jackson better than 50%. 4748
Collins case (40.5%): the newest Maine poll puts the incumbent three points ahead, and that release is where the board's 10-point move came from. 47
Conditional trade inference: all three election markets in this edition are priced off the same habit — the crowd follows whichever poll is newest and largest. Brazil tests that habit hardest, because three polls there agree on a tie and the board pays 57.65% for one side anyway. For a reader, the practical question in each race is which poll the price has already absorbed: in Los Angeles the Berkeley survey is now in the number, in Maine the Siena survey is the one the move followed, and in Brazil the price sits 16 points above where three separate surveys leave the race. 91044

What to carry into the next data window

  • This week's US data run: job openings, the ADP employment report, the September PCE price index and the payroll report all land before the weekend, and each one moves the October Fed contract from where it sits at 65.5%. 17
  • September 30 expiries: the September ceasefire contract, all three blockade contracts' nearest rungs, the September Bitcoin ladder, the September WTI ladder and the 30-year-yield contract all settle at the end of the month. 415
  • IMF PortWatch: the December Hormuz contract at 23.5% resolves the moment the fund publishes a seven-day moving average of transit calls at 60 or above, which makes its data page the fastest physical read on whether the strait is coming back. 6
  • October 1 and November 3: Los Angeles mail ballots go out on October 1, and the runoff plus the midterm elections fall on November 3, the date Trump named as the window for a US-Iran deal. 2343
  • October 4 and October 25: Brazil's first round on October 4 is the next hard fact for the widest poll-versus-market gap on the board, with the runoff on October 25. 37
  • FOMC: October 27-28, then December 8-9 with a Summary of Economic Projections, which is where the December contract's 78% for a quarter-point increase gets tested. 212

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
    Polymarket: Fed Decision in October?

    gamma-api.polymarket.com

  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48

This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.

Related content