Fri: October payrolls printed +165,000, past the firm line, and the 5- and 10-year closed higher by 4 and 2 basis points after dipping below Thursday's close at midday. Two hits of two.
Written Friday November 6, 2026, at about 6:00 PM ET, covering Friday November 6. Nonfarm payrolls rose 165,000 on the first estimate (BLS), with August and September revised up a combined 84,000 and unemployment at 4.1 percent, so the Firm row triggered: the 5-year closed at 4.648 percent (Thursday 4.608) and the 10-year at 4.859 (4.837) (Yahoo), hits by 4.0 and 2.2 basis points after both printed below Thursday's close in the 12:20 PM bar. Most of the rise was in place before 8:30. S&P 500 up 0.31 percent, dollar index up 0.7 percent, gold down 0.8 percent. CFTC: leveraged funds added 237,656 to the 5-year short, the first Treasury change past its noise floor in weeks. Bowman and G.19 recorded, not graded.
Written Friday November 6, 2026, at about 6:00 PM ET, covering Friday November 6's session. The plan graded below (editorial/plans/2026-11-06.md) was committed Thursday night at 8:54 PM ET, before any Friday US release; its own timing paragraph says about 9:00 PM. The Monday plan was committed tonight at 5:55 PM ET, before this note. October nonfarm payrolls printed +165,000 on the first estimate, past the plan's firm line of +100,000, so the Firm row triggered: the 5-year closed at 4.648 percent against Thursday's 4.608 and the 10-year at 4.859 against 4.837. Two hits of two, by 4.0 and 2.2 basis points. Both lines spent part of the early afternoon on the miss side: the 5-year's low was 4.601 and the 10-year's 4.814, both below Thursday's close, before the last two hours brought them back. The plan grades the close, so the grade is two hits, and the intraday dip is recorded here so that nobody has to take the number on trust.
This continues Thursday's note, where claims printed inside their band on a day the S&P 500 fell 4.36 percent and the 2-year yield fell 12 basis points. Thursday's closes, 4.608 and 4.837 percent, are the levels this plan graded against, and the Friday plan said in advance that Thursday's selloff and whatever followed it would sit inside the grade, not be adjusted for.
What the data said
- BLS Employment Situation, October 2026, 8:30 AM ET. Total nonfarm payroll employment rose by 165,000 in October; the unemployment rate was 4.1 percent, from 4.2 in September. The change for August was revised up from +133,000 to +186,000 and the change for September from +29,000 to +60,000, a combined +84,000. Average hourly earnings rose 14 cents, or 0.4 percent, to $38.00, and 3.1 percent over the past 12 months (BLS, The Employment Situation, October 2026, fetched 5:50 PM ET). The FRED vintage after the release (series PAYEMS, via FREDSTER, 5:50 PM ET) has October at 159,293 thousand, September 159,128, August 159,068 and July 158,882, which reproduces the three monthly changes. The next release, for November, is Friday December 4 at 8:30 AM ET (same BLS page). Payrolls graded below; the unemployment rate, earnings and revisions recorded, not graded. The paired reference, payroll revisions, works through why the first print is the one the plan fixes its line on, and what the +84,000 does and does not say.
- What the market expected, secondhand. Reuters (Lucia Mutikani and others, via Yahoo Finance, published 9:45 AM ET Nov 6) gave the Reuters poll consensus as +65,000 and wrote that "bets on a Fed hike next month rose to 78 percent from 70 percent". That percentage is Reuters' own rate-expectations figure; this site read no CME page and computes none. The direction matters more than the number: the pricing being discussed is a December hike, which is consistent with the Bank of England's quarter-point increase on Thursday, and not a cut. Reuters' intraday snapshot, without a timestamp, had the 2-year up 6 basis points at 4.52 percent, the 10-year up 3 at 4.87 and the 30-year up 2 at 5.27.
- Vice Chair for Supervision Bowman, 12:15 PM ET, "Supervision and Regulation", University of Mississippi. No text was published on federalreserve.gov at writing (the speech URL returns a 404) and a search found no coverage. Topic supervision; recorded, not graded.
- G.19 consumer credit, 3:00 PM ET. "Consumer credit increased at a seasonally adjusted annual rate of 4.2 percent during the third quarter," and "In September, consumer credit increased at an annual rate of 6.4 percent" (Federal Reserve, G.19, release date November 6, 2026). Recorded, not graded.
- CFTC Traders in Financial Futures, 3:30 PM ET, positions as of Tuesday November 3. In the positioning section below. Recorded, not graded.
What moved, and when
The 5-year (^FVX) closed at 4.648 percent from 4.608, up 4.0 basis points, on a daily bar that opened 4.643, ran to 4.679 and bottomed at 4.601; the 10-year (^TNX) closed 4.859 from 4.837, up 2.2, on a bar of 4.874 open, 4.903 high, 4.814 low (Yahoo daily bars, pulled about 5:45 PM ET; yield bars close 2:59 PM ET). Most of the rise was in place before the release: at the 8:20 AM bar open, ten minutes before the print, ^FVX was 4.643 and ^TNX 4.874, already 3.5 and 3.7 basis points above Thursday's closes. The release window itself, 8:20 bar open to 8:50 bar open, added 1.4 basis points to the 5-year (4.643 to 4.657, with a high of 4.679) and took 1.0 off the 10-year (4.874 to 4.864, high 4.903); the 9:20 bars closed 4.671 and 4.882, the session's highest closes on the 30-minute series. Then both faded. The 12:20 bar on ^FVX opened 4.645, printed the day's low of 4.601 and closed 4.620; the same bar on ^TNX printed 4.814 and closed 4.835. For part of that half hour both lines were below Thursday's close, which is the miss side of the plan's table. The 2:50 bars closed 4.648 and 4.859, the daily closes. The 12:20 bar spans Bowman's 12:15 PM discussion; the topic was supervision, no text was published, and this note records the coincidence without attributing the move to it. Event windows explains why the window is recorded and never graded.
On Treasury's closing par curve the move was led by the front end and the belly: the 1-year rose 5 basis points to 4.33 percent, the 2-year 5 to 4.51, the 3-year 6 to 4.59, the 5-year 5 to 4.64, the 7-year 4 to 4.74, the 10-year 2 to 4.86, the 20-year 2 to 5.29 and the 30-year 1 to 5.26; the full Nov 6 curve runs 1-month 4.02, 2-month 4.15, 3-month 4.19, 6-month 4.23, 1-year 4.33, 2-year 4.51, 3-year 4.59, 5-year 4.64, 7-year 4.74, 10-year 4.86, 20-year 5.29, 30-year 5.26, against Nov 5's 1-month 4.03, 2-month 4.15, 3-month 4.17, 6-month 4.19, 1-year 4.28, 2-year 4.46, 3-year 4.53, 5-year 4.59, 7-year 4.70, 10-year 4.84, 20-year 5.27, 30-year 5.25 (Treasury nominal curve, fetched 5:45 PM ET). The 2s10s spread narrowed from 38 to 35 basis points. Thursday's rally was front-end-led and so was Friday's reversal of part of it: the 2-year gave back 5 of Thursday's 12, the 10-year 2 of its 6. The 3-year point, 4.59, is the reference the Monday plan records alongside its 3-year auction leg.
Equities finished higher and well off the high. The S&P 500 (^GSPC) closed at 7,376.86 from 7,353.73, up 0.31 percent, on a bar that opened 7,377.65, reached 7,444.20 and bottomed at 7,369.55, so the close was within 8 points of the low after a high that was 1.2 percent up (Yahoo). CNBC's closing coverage gives the S&P 500 at 7,376.86, matching Yahoo, the Nasdaq Composite up 1.1 percent to about 24,579 and the Dow down 0.5 percent to about 51,285, and puts the week at minus 2.7 percent for the S&P 500, minus 3.4 percent for the Nasdaq and minus 0.9 percent for the Dow (CNBC, Nov 6, 2026; secondhand, not read from an exchange). ES=F closed 7,404.0 from 7,383.0.
The dollar rose and gold fell. The dollar index (DX-Y.NYB) closed 102.279 from 101.55, up 0.7 percent, on a bar of 101.494 open, 102.358 high, 101.553 low; gold (GC=F) closed 4,079.5 from 4,113.5, down 0.8 percent (both Yahoo daily bars, re-pulled 5:52 PM ET with the last trade stamped 4:59 PM). January Brent (BZF27.NYM) closed 97.76 from 98.29. USD/JPY was 158.425, EUR/USD 1.11433 and GBP/USD 1.33558 on the Yahoo bars that opened Thursday at 7:00 PM ET and cover Friday's session, against 158.357 on the corresponding Thursday bar for the yen. None of that is graded. One convention note: Thursday's note quoted DXY at 101.62 and gold at 4,082.0 as live 8:38 PM prints while their daily bars were still trading; the completed Thursday bars are 101.55 and 4,113.5, which is why the Thursday column below differs from Thursday's note.
The tape
| Market | Thu Nov 5 | Fri Nov 6 | Change |
|---|---|---|---|
| 5Y yield (^FVX close) | 4.608% | 4.648% | +4.0 bp |
| 10Y yield (^TNX close) | 4.837% | 4.859% | +2.2 bp |
| 30Y yield (^TYX close) | 5.251% | 5.260% | +0.9 bp |
| 13-week (^IRX close) | 3.987% | 4.000% | +1.3 bp |
| S&P 500 (^GSPC close) | 7,353.73 | 7,376.86 | +0.31% |
| Brent, January (BZF27 daily close) | 98.29 | 97.76 | -0.53 (-0.5%) |
| Gold (GC=F daily close) | 4,113.5 | 4,079.5 | -0.8% |
| DXY (DX-Y.NYB daily close) | 101.55 | 102.279 | +0.7% |
| USD/JPY (bar opened 7 PM ET prior evening) | 158.357 | 158.425 | +0.04% |
| ES=F (daily close) | 7,383.0 | 7,404.0 | +0.3% |
Sources: Yahoo Finance query1 (v8/finance/chart), daily bars, pulled about 5:45 PM ET Friday; DXY and gold re-pulled at 5:52 PM ET as completed bars. Every row is a completed daily bar. Treasury's curve and Yahoo agree on direction and roughly on size at the 5- and 10-year.
Grading the plan
Leg A, October nonfarm payrolls, monthly change, seasonally adjusted, as first printed by the BLS at 8:30 AM ET. At zero or below, the 5- and 10-year close below Thursday's 4.608 and 4.837 percent; at +100,000 or above, above them; +1,000 to +99,000, no call. Classification: +165,000, firm.
| Line in the map | Expected | Actual | Grade |
|---|---|---|---|
| Soft (≤ 0): ^FVX below 4.608%, ^TNX below 4.837% | if triggered | payrolls +165,000 | Not triggered |
| Firm (≥ +100,000): ^FVX above 4.608% | above 4.608% | 4.648% | Hit |
| Firm (≥ +100,000): ^TNX above 4.837% | above 4.837% | 4.859% | Hit |
| Neither (+1,000 to +99,000) | no call | payrolls +165,000 | Not triggered |
Grade: two hits, no misses, no no-calls, nothing void. The plan set its lines from the full spread of the last three first prints, minus 23,000 for July, plus 162,000 for August and plus 29,000 for September, and said it accepted a no call anywhere from +1,000 to +99,000. October's +165,000 cleared the firm line by 65,000, and the Reuters consensus of +65,000 would have been a no call by the plan's own table, so the print was firm against the plan and firm against the poll. The yields did what the firm row said, and the honest qualifiers are three. First, the margins were 4.0 and 2.2 basis points, and the 10-year's is small enough that a different closing print would have flipped it; the Oct 27 note called a 0.7 basis point hit "real by the plan's terms and small enough to be noise", and 2.2 is closer to that than to a clean move. Second, both lines were on the miss side for part of the 12:20 bar, so a plan graded at 12:30 PM would have scored two misses on the same print; the plan grades the close, which was fixed before the day, and the close is what is graded. Third, most of the day's rise was in place at 8:20 AM, before the print, so the window attributes about a third of the 5-year's close-to-close move to the release and none of the 10-year's. The record takes the two hits; this paragraph is what they are worth.
This is the first plan since November 3 to trigger a row at all, after three consecutive no-call days (ISM services 54.5, claims 189,000, and before them the ECI at 0.8 percent). The payrolls band was built wide because the series' last three first prints spanned 185,000, and this time the print was outside the band on the firm side. The no-call band is the reference for why a band set at a series' recent range only grades the series when it leaves that range, and today it did.
Positioning, for the record
The CFTC Traders in Financial Futures report released at 3:30 PM ET (futures only, positions as of Tuesday November 3) has leveraged funds net short 2,315,516 contracts of the 5-year note, 237,656 more than a week earlier, and net short 2,262,310 of the 10-year, 35,035 more; asset managers are net long 3,177,655 and 3,030,236, the 5-year long up 238,754 (bin/tff_report.py --weeks 52, run 5:54 PM ET). The 5-year change is the first Treasury change in several weeks to clear the script's noise floor, at about twice the 115,599 threshold, and it moved the 5-year short from the 90th percentile of the trailing 52 weeks (near the year's smallest short) to the 54th. The 2-year change, minus 93,234, sits just inside its 101,282 floor and is written as noise; the 10-year and bond changes were well inside theirs. The bond short, 166,257, is no longer the window's smallest. The euro leveraged net, minus 78,559, remains the window's most short at the 1.9th percentile; the yen net changed by 187 contracts. These positions were struck Tuesday, before Thursday's rally and Friday's reversal, and say nothing about either. The next report, for November 10 positions, is released Monday November 16 rather than Friday November 13: the CFTC's schedule marks the November 16 release as delayed by Veterans Day (fetched 5:50 PM ET). The Traders in Financial Futures guide is the reference.
What we do not know
- What drove the midday fade. Both yields fell to below Thursday's close in the 12:20 bar and recovered by the close. Bowman spoke at 12:15 on supervision; no text was published and nothing attributes the move to her. No other sourced cause was read.
- What the December pricing is. Reuters' 78 percent is Reuters' figure, direction hike, and this site has not read the CME page. The direction of the 2-year (up 5 basis points on Treasury's curve, taking back less than half of Thursday's fall) is the evidence; the odds are not.
- Why the dollar rose 0.7 percent and gold fell 0.8 percent. Consistent with a firm print and higher front-end yields, but no sourced explanation was read and none is supplied.
- The Nasdaq and Dow closes, and the weekly changes. CNBC's figures, quoted as approximate and secondhand.
- Whether the +84,000 upward revision to August and September says anything about November. The paired reference says why it does not.
What comes next
A plan for Monday November 9 (editorial/plans/2026-11-09.md) was committed at 5:55 PM ET tonight, before this note and before any Monday US event. The only Monday catalyst with a number is the $58 billion 3-year note auction at 1:00 PM ET (TreasuryDirect upcoming feed; new issue, CUSIP 91282CRW3), and the plan grades one line on it: weak at a bid-to-cover of 2.53 or below or an indirect share of 57.5 percent or below, strong at 2.73 or above and 67.6 percent or above, thresholds set just outside the six-auction range since May, with the 5-year graded against tonight's 4.648 percent. Bowman's 8:40 AM discussion on community banking and Treasury's 3-year par point are recorded, not graded. The rest of the week, from the BLS schedule, the Fed Board calendar and TreasuryDirect: Tuesday November 10, Waller at 4:00 AM ET in London, October CPI at 8:30 AM ET, the 10-year auction at 1:00 PM and Bowman on the outlook at 3:30 and 6:30 PM; Wednesday November 11, Veterans Day, with the Fed's statistical releases moved to Thursday, the dollar bond market closed by SIFMA's recommendation and the NYSE open by its own holiday list; Thursday November 12, claims by the weekly convention, Barr at 10:00 AM and the 30-year auction at 1:00 PM; Friday November 13, October PPI at 8:30 AM ET, Bowman at 10:50 AM and Jefferson at 7:00 PM. The record since mid-October: Oct 14 two hits, two misses; Oct 16 no hits, two misses; Oct 21, 22 and 23 nothing triggered; Oct 27 one hit; Oct 30 nothing triggered; Nov 3 two hits; Nov 4 nothing triggered; Nov 5 nothing triggered; Nov 6 two hits. No plans for Oct 28, Oct 29 or Nov 2. The full table is on the scorecard.
What to read alongside
- Payroll revisions (the paired reference)
- The NFP report
- The no-call band
- Flight to quality (Thursday's reference, for the rally this session partly reversed)
- Event windows
Nothing on this site is investment advice or a recommendation to trade. Most retail traders lose money. Conditional statements published here are falsifiable hypotheses, graded in public, not signals.