The Friday Wrap — August 7, 2026: a jobs miss, and the best week in months

The economy lost jobs in July, and stocks had their best week in months. Those two facts belong in the same sentence, because the second one happened because of the first.

Nonfarm payrolls fell 23,000 against an expected gain of about 83,000, per the Bureau of Labor Statistics release Friday morning. Private payrolls actually grew 30,000; government shed 53,000, with additional softness in retail and in leisure and hospitality. Revisions took 103,000 jobs out of May and June. The unemployment rate went down to 4.1% — but on a shrinking labor force, not on hiring. Average hourly earnings growth slipped to 3.2% over twelve months, the slowest since May 2021.

For most of 2026 a weak labor print would have been a problem. This year it is a release valve, because the pressure has been coming from the other side.

Monetary — the committee that wanted to hike

At the July 29 meeting the FOMC held the target range at 3.50%–3.75%. The vote was 9–3. All three dissents ran in the same direction, and it was not the direction most of the last decade trained people to expect: the statement recorded that the dissenters preferred to raise the range by a quarter point at that meeting. Three regional Reserve Bank presidents wanted tighter policy, now.

That is the frame the jobs report landed into. Coming into Friday, futures had a coin-flip-or-worse on a September hike; after the print they did not. CNBC's read of fed funds futures put the odds of a hold at 60%, up from 45% Thursday. Reuters' read had the probability of tightening at 43.9%, down from 57% pre-print. The two numbers are not identical and we are not going to average them — different contracts, different snapshot times. What they agree on is the size and direction of the move: a market that spent the week bracing for tighter policy stopped bracing on Friday morning.

Treasury yields fell on the print. Our proxies caught it: TLT (20y+ Treasuries) closed +0.29% Friday and +1.03% on the week, IEF (7–10y) +0.24% and +0.58%.

Earnings — the tape sorted, hard

Roughly three-quarters of the S&P 500 has now reported. Coverage puts second-quarter growth in the high-20s to low-30s percent year over year — FactSet's count via CNBC had about 300 companies in with 85% beating and index earnings tracking +29%; other reads put adjusted growth near +31%, which would be the strongest since 2021. Take the range, not a point estimate.

The more useful thing is how the tape treated the prints, and this week it treated them with almost no mercy in either direction. We checked each against broker daily bars rather than quoting the coverage, because after-hours moves and regular sessions are different numbers and get mixed freely.

Palantir reported Monday after the close: U.S. commercial revenue up 149% year over year to $764 million, with full-year U.S. commercial guidance raised to above $3.42 billion from $3.22 billion. Headlines led with a 12% after-hours pop. The actual regular session Tuesday ran 125.65 to 162.66 — +29.45%, on 175 million shares against a recent daily norm near 27 million. It gave back 2.6% Wednesday and 1.6% Thursday and kept nearly all of it.

AMD reported Tuesday after the close: adjusted EPS $1.66 against $1.61 expected, a beat — alongside a quarter-over-quarter decline in free cash flow margin. Coverage led with −8.8% after hours. The regular session Wednesday was 518.58 to 482.05, −7.04%, then +1.50% Thursday. An EPS beat, and the tape sold the cash-flow line.

Disney reported Wednesday before the open and gained +3.65% in that session (98.18 to 101.76), adding another +2.87% Thursday.

Same week, same market: +29% for the company that showed conversion, −7% for the company that beat on EPS while its cash generation slipped. That sorting is the story, and it is why an index-level "earnings were strong" summary tells you almost nothing about what you owned.

Energy and geopolitics — the war premium came out

Crude had its worst week in some time. Our WTI proxy USO closed at 117.98, −8.66% on the week and −8.66% month-to-date. Brent was quoted near $82 Friday; WTI traded under $80, with one session accounting for roughly a $5.89 drop.

The coverage attributes it to three things arriving together. Iran and world powers agreed to resume nuclear talks this month, which markets read as a path toward sanctions relief and more barrels. The U.S. president claimed a deal had been reached over the Strait of Hormuz — that claim is unconfirmed and Iran has disputed the account, and we are reporting it as a claim, not a fact. And OPEC+ agreed to raise September production by 188,000 barrels per day, completing the rollback of the voluntary cuts put in place in 2023.

Note what the longer frame still says: USO is −8.66% this month and still +16.29% against its 200-day average. The premium that came out this week had been built over months. Both things are true at once.

Gold went the other way and went hard. GLD closed 398.47, +7.25% on the week, its strongest weekly gain since January per the coverage (which puts spot gold's move near +6.6% and above $4,300/oz — ours is the ETF, theirs is spot). Falling yields, a softer labor market, and drained energy risk all pushed the same direction. Worth keeping honest, though: GLD is still −3.21% against its own 200-day. A great week does not undo the trend it sits inside.

The dollar proxy barely moved — UUP −0.43% Friday, −0.35% on the week, still +1.96% versus its 200-day. Coverage describing a "sliding dollar" is directionally consistent with our bars but considerably louder than what a third of a percent looks like.

Calendar periods — August is repaying July

| | Aug MTD | July | QTD | YTD | |---|---|---|---|---| | SPY | +3.51% | +0.03% | +3.55% | +14.00% | | QQQ | +5.09% | −6.57% | −1.82% | +17.98% | | IWM | +3.56% | −3.08% | +0.37% | +23.02% |

Five sessions into August, the thing that was worst in July is best. QQQ lost 6.57% in July and has taken 5.09% back in a week — though it is still −1.82% for the quarter, which is the honest way to hold both. Dispersion this month is 1.58 percentage points between QQQ and SPY, far narrower than July's 6.60-point split. The month so far has been broad rather than concentrated.

Zoom out and none of this looks like distress. All three are up double digits on the year, and the leader is the small-cap index at +23.02%, ahead of QQQ's +17.98% and SPY's +14.00%.

The weekly frame

SPY closed 773.26, +3.51% on the week, ending any red-week streak at zero. It sits +3.74% above the 10-week SMA (745.35) and +3.28% above the 8-week EMA (748.71). Weekly MACD histogram is positive at +0.80, weekly RSI 67.6, weekly stochastic K 94.4 / D 79.6 — the top of the stochastic band, which describes where price is, not where it goes.

QQQ is the interesting one: +5.09% on the week and back above both weekly averages (+1.84% vs the 10-week, +2.57% vs the 8-week) — yet its weekly MACD histogram is still negative at −3.10. The price recovered faster than the weekly momentum series, which has not turned. IWM shows a milder version of the same: +3.56% on the week, above both weekly averages, weekly MACD histogram −0.11.

One thing we track did not fire: our weekly analog condition (two-plus red weeks and below both the 10-week SMA and the 8-week EMA) is not in effect this week. It carries nothing here, and we are not going to dress it up.

The daily tape

SPY: Friday's range was 769.61–773.91, closing 773.26 above its open. The daily stack is fully beneath price — 10-day 753.87, 20-day 750.17, 50-day 746.61, 200-day 700.11 (+10.45%). The 20/50 spread is +0.48%, widening by 0.056 points. Daily MACD histogram +3.29, RSI 66.0, stochastic K 92.5 / D 86.8. ATR(14) is 7.86, about 1.02% of price. The 60-day range is 716.58–776.85 and the 20-day range is 729.10–776.85 — meaning Friday's close is 0.46% below the 60-day high, and both range highs are the same print.

That matters, because several summaries this week described record highs across the major indices. Our own bars do not support that for these ETFs: SPY closed under its 60-day high, and QQQ at 723.03 sits 3.3% below its 60-day high of 747.83. We flag the disagreement rather than pick a side, but every price in this letter comes from our SIP bars, and those are the ones we will defend.

QQQ's daily structure is genuinely different: price is above the 10-day (696.98) and 20-day (700.34) and now +1.22% above the 50-day (714.31), but the 20/50 spread is still negative at −1.96% — the shorter average has not crossed back over the longer one. Daily MACD line just crossed to +0.17 with a +4.42 histogram; RSI 57.2; ATR 12.29, 1.70% of price, notably wider than SPY's. Friday's candle carried a long lower tail — 51% of the day's range below the body — with the close above the open.

IWM: 301.56, above all four averages, 50-day +2.67%, 200-day +13.25%, daily MACD histogram +0.75, RSI 59.7, ATR 3.93 (1.30%). Its 20-day high of 303.06 is also the 60-day high.

Breadth

The three indices disagree in a specific and readable way. On the year, small caps lead by nine points over SPY. On the month, the Nasdaq proxy leads. But on structure, QQQ is the only one of the three with a negative 20/50 spread — it is participating in the rally without yet having repaired the shorter-term damage July did. SPY and IWM both have clean positive stacks. RSI tells the same story from a different angle: SPY 66.0 and IWM 59.7 against QQQ's 57.2, even though QQQ had the biggest week.

Open interest

Strikes and contract counts only — our data key carries no greeks, so this is open interest, never gamma. OCC as-of 2026-08-05, which lags and is not Friday's book.

The August 7 expiry cleared this week with call OI concentrated at 775 (19,587) and 760 (15,914), and much heavier put OI at 750 (49,969) and 730 (45,838). For the next expiry, August 10: largest call OI at 780 (5,075) and 800 (4,950); largest put OI at 750 (11,711) and 729 (11,625). Spot closed 773.26, between them. Structurally, the nearest large call line above is 780 and the next is 800; below spot, the nearest comparable put line is 750, and beneath that 729 — a wide gap with nothing of similar size between.

Next week

Our local audited econ list expired July 29, so the macro dates below come from this week's research, not from that file: July CPI is scheduled for August 12, with PPI and retail sales later in the week. Coverage weekday labels were inconsistent, so treat the dates as scheduled and expect them to move.

From our local earnings cache — semis only, 13 symbols, fetched June 29 — the one entry in the next nine days is AMAT on August 13. That is not the week's full slate; it is the slice we cache.

What we don't carry

Same list as always, and we would rather name it than fake it: no dark-pool prints, no market-on-close imbalance, no spot VIX / VIX1D / VVIX, no true GEX or greeks, no FOMC probability feed of our own. VIXY is a VIX futures ETF, not spot — it closed 19.56, −4.63% on the week and −28.87% against its 200-day. USO is not WTI and UUP is not DXY. Where we used an outside number this week, it is attributed above.

From the Coil board

As of August 7, the account is in SWING mode, and the board is more cautious than the tape. SPX reads RISK_OFF — cash, with SPY 3.6% extended above its 50-day and no qualifying setup beneath it. QQQ is the one lane with anything open: NAMES_ON, chop-capped, with one of six sectors open (Biotech & Pharma) and 42 names fully qualified. Macro is RISK_OFF and defensive. Crypto is gated off entirely — BTC closed 8.9% below its 200-day gate, so both sleeves sit in cash until a UTC daily close reopens it.

A board that says "cash" through a +3.5% week is worth saying out loud rather than hiding: our rules want a setup, not a rally already in progress.

---

Read the live board at coil.trade/scanner · own the engine, $29 at coil.trade.

The Friday Wrap is impersonal research and market commentary. It is not investment advice, not a recommendation, and not personalized to anyone's situation.

Prices, ranges, moving averages, and index moves: our own Alpaca SIP daily and weekly bars. Single-stock earnings sessions verified against broker daily bars. Open interest: OCC, as of 2026-08-05. Macro, policy, energy, and calendar reporting drawn from CNBC, Reuters via Yahoo Finance, the BLS employment release, the Federal Reserve's July 29 statement, Quartz, TheStreet, Mining.com, and FXLeaders.