2026-10-23 16:00 ET — Consumer staples (cluster 4) after the Oct 22 defensive spike: PG/PM prints, KO (Tue) and MO/MDLZ (Thu) next week

Slot: Fri 2026-10-23, 16:00 ET hourly research (written just after the close). Status: FINAL (§9 whiggity consult added ~16:50 ET).

TRAI data used (all returned 200 at ~16:05 ET):

• /api/latest: run 2026-10-22_premarket_2026-10-23T12-53-19-641Z (8:53 AM ET; daily stats = Oct 22 session).
• /api/events (asOf 2026-10-23, 5-trading-day gap-risk window), /api/correlations (60 sessions to 2026-10-22), /api/scorecard (replay 2026-04-23 → 2026-10-21), /api/macro (asOf 2026-10-22), /api/regime, /api/options.

Other data (my pulls and calcs): Yahoo chart API daily bars (2022-01 → 2026-10-23 close; regularMarketTime 16:00 ET), earnings-date history from Yahoo via yfinance. "%" = vs the Oct 22 close unless stated. "o→c" = today's open to close.

1. Focus and why

• Focus: TRAI correlation cluster 4 — PM, MO, KO, PEP, MDLZ, PG, KHC (avg ρ 0.567, all Consumer Staples), plus WMT/COST (cluster 26) and TGT as context.
• Why this hour:
• Three of today's 8 mean_reversion fade-shorts are staples (MO, PEP, TGT), after XLP's +3.31% on Oct 22, its second-biggest day since 2022 (after +3.88% on 2025-04-09; Yahoo, my calc). PG (+5.58%, residualZ 5.35, own print) and KHC (+6.77%, z 3.75) had bigger residuals but fail TRAI liquidity (PG ATR% 1.90% < 2%; KHC ADV $327M < $500M), so they are not in a bucket.
• Staples led today's S&P declines (Reuters, below), and the fades all worked into the close.
• Next week is the cluster's heaviest print week: KO Tue 10-27 BMO, MO Thu 10-29 BMO, MDLZ Thu 10-29 AMC (reaction Fri 10-30), CL Fri 10-30 BMO (not in universe), KHC Wed 11-04 BMO. KO and MO are both inside TRAI's 5-day "elevated" window now.
• Rotation: today's earlier reports covered semis (09:00), software (10:00), telecom/energy (11:00), AI power/industrials (12:00), payments (13:00), mega-cap cloud (14:00) and alt managers/SPOT (15:00). The 10:00 report covered HD/LOW/TGT as a rates trade; staples (PG/KO/PEP/MO/PM/MDLZ) have not been covered.
• One-line read: Oct 22's staples spike was a rotation out of semis plus PG's beat; today it half-reversed (XLP −1.92%) on a hot-oil/hot-inflation tape with SMH up. Next week, the cluster's gap risk is KO on Tue (0/4 gaps filled, avg 2.32 ATR) and MO on Thu (0/4 filled, avg 2.73 ATR, 3 of last 4 down ≥ 5%). Nothing in the last hour changes Monday's gap risk on a candidate.

2. What the businesses are and what moves them

• Household & personal care (PG; CL out of universe): branded detergents, paper, beauty. Moves on organic sales (volume vs price/mix), commodity/freight costs and FX.
• Beverages & snacks (KO, PEP, MDLZ, KHC; KDP out of universe): KO is mostly concentrate sales to bottlers (high margin, less commodity exposure than PEP's snacks-heavy mix). PEP and MDLZ carry more input-cost (cocoa, grains, packaging, fuel) exposure.
• Tobacco/nicotine (PM, MO): PM is international with a growing smoke-free segment (IQOS, ZYN); MO is US-only cigarettes plus smoke-free. Pricing power and volume declines drive MO; smoke-free growth and FX drive PM.
• Shared factor drivers in 2026:
• Anti-AI/semis rotation. Cluster-4 names' average 60d ρ to the core-semis basket (NVDA, AMD, AVGO, MU, TSM, ASML, AMAT, LRCX, KLAC) is −0.38 (KHC) to −0.60 (MO) (TRAI /api/correlations, my calc). XLP–SMH daily-return ρ by year (Yahoo, my calc): 2022 +0.51, 2023 +0.19, 2024 −0.02, 2025 +0.01, 2026 YTD −0.39. Same sign-flip pattern as SPOT (S-049).
• Rates (bond-proxy): 10y at 4.96% Oct 22 (TRAI /api/macro, Treasury fallback); Yahoo ^TNX 4.99% today.
• Oil/energy costs: Brent closed just above $116 and WTI above $103 today (Yahoo Finance live, 2026-10-23 3:05 PM CDT). Both of this week's staples prints (PG, PM) flagged energy/commodity cost pressure (§3).
• Crowding: /api/macro (asOf 10-22): 20-day sector leaders vs SPY XLP +10.6%, XLU +9.3%, XLE +8.4%; XLP RS quadrant "leading".
• YTD to 10-23 close (Yahoo, my calc): KO +32.6%, MO +28.9%, PM +24.4%, CL +19.0%, MDLZ +18.5%, XLP +13.6%, PG +12.4%, SPY +11.4%, KHC +2.2%, PEP −5.4%; SMH +46.4%.

3. This week's prints and guidance (primary where possible)

3.1 PG — FY27 Q1, reported Thu 10-22 BMO

• Net sales $23.3B (+4%; organic +4% = volume +2, mix +1, pricing "minor"); diluted EPS $1.97, core EPS $2.00 (+1%); core operating margin −80 bp despite 250 bp productivity savings (P&G release, 2026-10-22).
• Guidance: all-in sales growth raised to 2–4% (Thorne acquisition, closed Oct 1); organic 1–3% maintained; core EPS $6.89–$7.11 maintained; commodity headwind raised to ~$1.4B after tax (from $1.0B per Reuters) (release above; Reuters via Investing.com, 2026-10-22). CFO: cost pressure to increase over the next two quarters.
• Consensus: core EPS $1.88, revenue $22.8B (LSEG, via CNBC, 2026-10-22).
• Reaction: 10-22 gap +3.38%, day-1 +2.13%, close-close +5.58% (TRAI /api/events). Day 2 (today): −2.70%, o→c −1.76%, framed as the cost hit ("PG at $161.25, −2.62% at 11:39 ET", GuruFocus, 2026-10-23; secondary).
• Next print: ~2027-01-21 (TRAI estimate).

3.2 PM — Q3, reported Wed 10-21 BMO

• Net revenue $11.4B (+5.5%, +7.1% organic); adj. diluted EPS $2.36 (+5.4%); reported $2.12; SFP >40% of revenue (PMI release, 2026-10-21).
• Guidance: adj. EPS $8.29–$8.44 kept, but PM now expects the middle-to-lower half, citing Iran-war energy costs (bigger Q4 impact), Ukraine disruption and higher US investment (Reuters via Investing.com, 2026-10-21).
• Reaction: gap −6.38%, then day-1 +5.51% (close-close −1.22%) (TRAI /api/events) — the gap was bought.
• Next print: not scheduled (TRAI next = null); historically early February.

3.3 PEP — Q3, reported 10-08 BMO (context)

• Net revenue $25.27B (+5.6%), organic +3.1%; core EPS growth guide cut to 2.5–3.5% from the low end of 5–7% (core constant-currency 1–2%, was the low end of 4–6%) on North America margin pressure: PFNA core margin −280 bp on affordability investments and A&M, while U.S. salty-snack volume grew (PDFs opened for §9 item 8) (PEP Q3 8-K Ex. 99.1; Investing.com transcript summary, 2026-10-08).
• Reaction 10-08: gap +0.47%, day-1 +3.24% (TRAI /api/events) despite the cut. PEP is the only cluster-4 name down YTD.

3.4 Today's news in the group

• KO: kept WPP Open X as global media partner ex-North America/Japan/South Korea; Omnicom picked for North America (MT Newswires via Yahoo, 2026-10-23 1:55 PM ET). Not an earnings catalyst.
• Same source: XLP −2.1% intraday; UMich October final sentiment 48, the lowest since May, with inflation expectations the highest since May.
• Wells Fargo note: diesel up ~60% y/y in Q3 hits discount-retailer margins (DLTR/DG −~5%) (CNBC live, 2026-10-23). Out of universe, but the same fuel-cost theme as PG/PM.
• Reuters: "consumer staples and financial groups leading the declines"; Dow −1.18%, S&P −0.43% (Reuters via Investing.com, 2026-10-23).
• No staples SEC filing, guidance change or company release found in the last hour (15:00–16:05 ET).

4. Today's tape (Yahoo daily, final closes)

\* Not in the TRAI universe. Oct 22 figures and signals: TRAI /api/latest. Oct 23: Yahoo daily.

• Live scoring of today's TRAI buckets (gross open→close, my calc; TRAI actually forbids entries before 10:00): mean_reversion fade-shorts (n 8) +1.31%, of which MO +1.89, PEP +1.62, TGT +2.66; fade-longs (n 9) +0.58%; long_momentum (n 6) +0.83%; short_momentum (n 12) −0.87%.
• Replay contrast: TRAI replay (net, my calc) for Thursday-signal mean_reversion fade-shorts: n 99, mean −0.92%, t −4.02; Thursday-signal fade-longs n 28, +0.89%, t 3.05. All MR fade-shorts: n 322, −0.26%, t −2.10. Staples fade-shorts: n 15, −0.69%, t −1.37. Today ran against the replay's strongest negative slice (→ S-054).
• ERRATUM (added 17:00 ET): the short-side replay figures in this bullet are sign-flipped (replayTrades[].oc is already signed; it was multiplied by side again). Corrected, net of 5 bp: Thursday-signal fade-shorts n 99, +0.76%, t +3.30; all MR fade-shorts n 322, +0.10%, t +0.77 (matches TRAI's ranking); staples fade-shorts n 15, +0.53%. Today's live result agreed with the replay. See 17:00 report §5.5 and S-054.

5. Does the semis direction decide the staples give-back? (Yahoo daily, 2022-01 → 2026-10-23, my calc)

• 11 "rotation days" with XLP ≥ +1.5% and SMH ≤ −3% (9 of them in 2026, including 10-14 and 10-22).
• Next day, when SMH rose (n 6: 2025-01-28, 2026-02-02, 06-08, 06-11, 07-06, 10-23): XLP c-c mean −0.67%, o→c −0.51%; six-name staples basket (MO, PEP, KO, PG, PM, MDLZ) o→c −1.03%.
• When SMH fell again (n 5): XLP c-c +0.47%, o→c −0.10%; basket o→c −0.43%.
• Unconditional, after any XLP ≥ +1.5% day (n 50): next-day XLP o→c mean −0.01%, median +0.14%. No unconditional reversal; the give-back shows up only when semis bounce. n is small; premarket SMH direction is an obvious filter to test (→ S-053).
• XLP's next-day gap after rotation days was small (−0.83% to +1.89%), so the move happens in-session, which suits TRAI's intraday holds.

6. Next week's cluster-4 catalysts and typical reactions

6.1 Calendar (verified)

6.2 KO into Tue 10-27

• Guide (from Q2, 2026-07-28): organic revenue ~5% (raised from 4–5%), comparable EPS +9–10%, comparable currency-neutral EPS ex-M&A +7–8%, FCF ~$12.4B. Q2 organic +6%, unit cases +5%, comparable EPS $0.97 (KO Q2 release, 2026-07-28).
• Consensus (secondary; three previews agree, §9 item 2): EPS ~$0.88, revenue ~$12.90–12.93B, FY26 EPS ~$3.30; "beaten in 8 of the last 8" (MarketBeat; Yahoo preview).
• TRAI last 4: avg |gap| 3.82% (2.32 ATR), fill 0/4, continuation 25%; gaps +3.11, −1.72, +5.41, +5.03 (TRAI /api/events). KO fails TRAI liquidity (ATR%), so the trade risk is via peers.
• KO print → peers (19 prints 2022–2026, Yahoo, my calc): same-sign rate PM 74%, PEP/KDP/MO/XLP 68%; |move| above baseline for PM 1.61% vs 1.11% and MDLZ 1.37% vs 1.00%; PEP 0.97% vs 0.94% (no bigger move). ρ(KO, peer) on print days: KDP 0.67, XLP 0.61, MDLZ 0.57, MO 0.49, PEP 0.44. KO's day-2 o→c mean +0.05% (no drift).

6.3 MO into Thu 10-29

• FY26 adj. EPS guide $5.61–$5.72; Q3 consensus $1.51 EPS / $5.31B revenue (Zacks $1.50 / $5.31B; §9 item 3 rejects whiggity's $1.42–1.45); Q2 $1.48 missed $1.50 (all MarketBeat, 2026-10-22; secondary).
• TRAI last 4: avg |gap| 5.42% (2.73 ATR), fill 0/4, continuation 75%, avg day-1 in gap direction +1.98%; reactions −7.81%, −5.34%, +6.52%, −9.32% close-close (TRAI /api/events). The only cluster-4 name whose gaps tend to extend, not reverse.
• MO print → peers (19 prints): only PM follows (same-sign 74%, ρ 0.54); KO/PG/KDP ρ ≈ 0. XLP |move| 0.87% vs 0.64% baseline.
• Same morning: GDP/PCE 8:30, ARES/OWL/TPG, MA/CAT BMO, and the GOOGL/MSFT reaction session (14:00 and 15:00 reports).

6.4 MDLZ (Thu 10-29 AMC → Fri 10-30) and CL (Fri 10-30 BMO)

• MDLZ last 4: avg |gap| 2.09% (0.97 ATR), continuation 75% (TRAI). Fails TRAI ADV, so peer-risk only; MDLZ–PEP ρ 0.68, MDLZ–PG 0.64.
• Fri 10-30 stacks MDLZ's reaction with CL's print. CL is out of universe; PG is its natural peer (not computed this hour).

6.5 Peer reactions to PG and PM prints (for reference)

• PG prints → XLP ρ 0.70, but only PEP/MDLZ follow (same-sign 60–65%); MO/PM do not.
• PM prints → MO only (same-sign 65%, ρ 0.43); PEP moves opposite (ρ −0.41).
• So cluster 4 is two sub-groups on print days: tobacco (PM↔MO) and food/beverage/HPC (KO, PEP, MDLZ, PG, KDP). The 60d cluster treats them as one.

7. Actionable for today / the next session (explicit)

• Today: none. TRAI is flat by 3:55 PM; the fade-shorts on MO/PEP/TGT worked (§4). No staples filing or guidance change after 15:00 ET.
• Monday 10-26: no cluster-4 print (VZ BMO is the only universe print, per /api/events). KOF (out of universe) now reports Mon BMO, a context read on bottler volumes/costs before KO (§9 item 9). The open risk is weekend oil/Iran headlines (Brent >$116), which hit staples through costs (PG/PM language) and flow through the semis-vs-defensives rotation (§5). Not a scheduled catalyst.
• Tue 10-27 (KO BMO) is the first staples gap-risk session: treat PEP, PM and MDLZ as peer-event names that morning (S-052). PM's follow is mostly the XLP factor, not a KO-specific link (§9 item 10). On Monday, a fade or momentum signal on PEP/PM/MO from Monday's data trades into KO's print day.
• Thu 10-29: MO is the higher-variance print (2.73 ATR, gaps extend). Do not fade MO's opening move that morning; PM is its only reliable follower.

8. Strategy implications (proposals; see strategy.md)

• S-052 (PROPOSAL, extends S-004/S-047): cluster-4 peer-event map with two sub-groups (tobacco: PM↔MO; food/bev/HPC: KO, PEP, MDLZ, PG). Flag only the sub-group on a print day. First cases: KO Tue 10-27 → PEP/PM/MDLZ; MO Thu 10-29 → PM.
• S-053 (NEEDS DATA): staples fade-shorts after a rotation day (XLP ≥ +1.5%, SMH ≤ −3%) only when SMH is up premarket. n 6 vs 5, basket o→c −1.03% vs −0.43%.
• S-054 (NEEDS DATA; figures corrected at 17:00 to n 99, +0.76%, t +3.30 — see erratum in §4): Thursday-signal → Friday mean_reversion fade-shorts: replay n 99, −0.92%, t −4.02 (the strongest negative slice found today), but today's live 8 fade-shorts made +1.31% gross. Check for a Friday/weekly-expiry effect, survivorship and the daily VWAP proxy before acting.
• S-049 addendum: XLP–SMH ρ flipped from +0.51 (2022) to −0.39 (2026 YTD), same pattern as SPOT.
• From the whiggity consult (§9): S-055 (day-2 staples reversion in high-oil regimes), S-056 (PM-follows-KO with DXY/10y/XLP controls; quick look 74% → 53%), S-057 (8:30 macro vs MO's opening gap; extends S-051), S-058 (gasoline vs the PM−MO earnings-week spread). All NEEDS DATA. S-052 now has a caveat: KO→PM is mostly sector beta.

9. Research partner (whiggity) consult

I consulted Kieran's Monica research bot whiggity after the close (chat) and gave it this report's draft summary. I treat its reply as input from a second researcher, not as a source. Items are tagged [whiggity]. A point is kept as verified only where I confirmed it against a public source I opened myself (link + date below; accessed 2026-10-23 ~16:20–16:50 ET) or against my own Yahoo/TRAI calcs. Everything else is listed as an open question and is not used as fact.

Verified (kept, with my citations)

1. [whiggity] "PG's Oct 22 report was fiscal Q1 FY2027, not Q3": correct, and this report already says so. The release is titled "P&G Announces Fiscal Year 2027 First Quarter Results" and covers July–September 2026 (P&G release, 2026-10-22). I searched the report, the PG/sector notes and strategy.md: none of them labels it "Q3", so no fix was needed. The release also confirms the §3.1 figures (sales $23.3B, core EPS $2.00, commodity headwind ~$1.4B after tax "driven by higher raw materials, energy and transportation costs"; core gross margin −150 bp, of which −140 bp from commodities).
2. [whiggity] KO consensus. Revenue ~$12.90–12.93B: verified ($12.901B, Hudson Labs, 2026-10-06; $12.92B, MarketBeat, accessed 2026-10-23; $12.93B, 24/7 Wall St. via Yahoo, 2026-10-22). FY26 EPS ~$3.30: verified ($3.3032, same Yahoo article). Correction: Q3 comparable EPS consensus is $0.88 ($0.8810 MarketBeat/Yahoo; $0.88 Hudson Labs), not ~$0.87. Timing: "Oct. 27 before the NYSE opens", call 8:30 ET (KO IR, 2026-09-29). The specific 6:55 AM release time is not in KO's announcement: open question. I did not open the Perplexity page whiggity cited.
3. [whiggity] MO consensus. Revenue net of excise ~$5.31B: verified (S&P, 10 analysts, via moomoo, accessed 2026-10-23; Zacks via AInvest, 2026-10-14). Zacks Rank #2 (Buy): verified as of 2026-10-14 (same AInvest/Zacks piece). Stock ~$73–75: verified; MO closed $74.31 on 10-23 (moomoo; matches Yahoo, §4). Release ~7:00 ET on Thu 10-29, the same morning as GDP/PCE at 8:30: already verified in §6.1 (Altria IR, 2026-10-15).

Contradicted: whiggity's adj. EPS $1.42–1.45 is below every consensus I found: $1.51 (MarketBeat, 2026-10-22), $1.50 (Zacks via AInvest, 2026-10-14), $1.48 (Gotrade, which appears stale). $1.45 is the year-ago Q3 2025 actual on the Gotrade page, so whiggity likely mixed up the periods. I kept ~$1.50–1.51. The Barchart link returned 403 and was not read.

1. [whiggity] PM is >80% international: verified. In Q3 2026 the U.S. segment was $845M of $11,438M net revenue (7.4%), so ~93% is international (PMI Q3 release, SEC Ex. 99.1, 2026-10-21). MO is US-only (Altria 10-Q/8-K). The structural difference is real. Whether it explains the PM-follows-KO link is item 10.
2. [whiggity] Energy costs reach PG/PM margins with a lag: verified directionally for PG and PM, not for CL or the specific threshold. P&G's release blames the bigger commodity headwind on "raw materials, energy and transportation", and Reuters quotes the CFO saying cost pressure will increase over the next two quarters (§3.1). PMI's release cites "higher costs due to increases in energy-related input prices" (PMI release above). Open: the >$115 oil threshold, the "1–2 quarter lag" as a number, and anything about CL (it reports Fri 10-30; I have no CL source).
3. [whiggity] XLP is not a pure safe haven once oil spikes, and SMH acted as the hiding spot: consistent with today's tape, n = 1. On 10-23, with Brent above $116 (§2), XLP fell −1.92% and SMH rose +0.74% (§4). §5 already shows the staples give-back only appears when semis bounce. One day is not a regime test. The oil-threshold version is folded into S-055.

Partly right / contradicted by primary sources

1. [whiggity] PM's guide pinch read-through to MO: US ZYN volume vs promo intensity pressures MO's on!/on! PLUS ahead of 10-29. Facts verified (PMI release above): US ZYN shipments +4.2% to 3.2B pouches "despite challenging comparisons against a special promotional campaign last year"; US OCI −3.3% with "a significant increase in commercial investments"; PMI plans to "further increase investments" in Q4, including moving the ZYN flagship to 20-pouch cans with "transitional support programs". The MO baseline from its last print: on! shipments −4.2% y/y in Q2 2026 (49.9M cans); on! PLUS reached 120,000 stores, with national 12 mg expansion planned for Q3 and new flavors in Q4 (Altria Q2 release, SEC Ex. 99.1, 2026-07-30). The read-through itself is open: that PM's higher US spend will hurt on! in Q3 is an inference. Neither company says so.
2. [whiggity] PEP's 10-08 cut was driven by Frito-Lay North America volume and the low-income consumer: contradicted on volume. I opened the PDFs this time. PFNA had "an increase in volume and volume share in the U.S. savory and salty categories", with U.S. salty volume up four quarters in a row. PFNA core operating margin fell 280 bp on affordability investments, lapping an asset-sale gain, and higher A&M. PBNA organic volume declined. Management expects NA margin "to remain under pressure in Q4" (PEP Q3 prepared remarks, 2026-10-08; Q3 release). So the cut was a margin/price-investment story, not a volume collapse. "Low-income consumer" is a fair reading of "affordability", but the documents don't use that phrase. Guidance detail: core EPS growth 2.5–3.5% (was the low end of 5–7%) and core constant-currency 1–2% (was the low end of 4–6%). §3.3 now reflects this.
3. [whiggity] KO's refranchised bottler model shifts logistics/packaging inflation to bottlers; KOF also reports 10-27; CCEP linkage.
4. Model: verified. KO mainly sells concentrate to independent bottlers, who package and distribute (KO 2025 10-K, filed 2026-02-20). Hudson Labs lists CCBA refranchising as a margin tailwind for H2 2026 (above). That it shields KO from oil/packaging inflation this quarter is plausible but not quantified: open.
5. KOF date: wrong. On 2026-10-21 Coca-Cola FEMSA moved its Q3 release to Mon 10-26 before the open (call 11:00 ET) from 10-27 (KOF IR notice, 2026-10-21). MarketBeat still lists 10-27, which is probably where whiggity got it. KOF is out of TRAI's universe, but it is now a Monday pre-KO read on bottler volumes/costs (Mexico, Brazil). Added to §6.1.
6. CCEP: its Q3 trading update is Tue 11-03, 07:00 GMT (CCEP Q3 2026 trading update page), after KO. No KO-day linkage. Out of universe.
7. [whiggity] Counter-evidence: "PM-follows-KO 74%" may be a common USD/DXY + macro factor, and the PM/MO tobacco factor can break on FX or gasoline spikes. Quick test (Yahoo daily 2022-01 → 2026-10, my calc, n = 19 KO prints). I regressed KO and PM daily returns on XLP, DXY (DX-Y.NYB) and Δ10y (^TNX) using non-KO-print days, then compared residual signs on KO print days:
8. Raw same-sign: 74% (14/19).
9. After removing DXY + 10y only: 79%. The dollar does not explain the link (PM's DXY beta −0.33, ρ −0.22).
10. After also removing XLP: 53% (10/19), roughly a coin flip. Residual ρ 0.21 vs 0.30 raw.
11. 3 of the 19 KO dates were also PM's own print days (2022-02-10, 2025-07-22, 2025-10-21). Excluding them: raw 75%, residual 50% (n = 16).

So whiggity's ~50% guess holds, but the mechanism is the staples sector factor (XLP), not the dollar. S-052's KO→PM flag is a sector-beta flag, not a KO-specific read-through. The "tobacco factor breaks on FX/gasoline" claim was not tested; it is filed as S-058.

Open questions (not verified; not used as facts)

• [whiggity] Marlboro volume run-rate −9% to −11%. My source disagrees: Altria's last print showed Marlboro shipments −7.4% in Q2 2026 and −7.6% in H1; total domestic cigarettes −3.2% (−4.5% adjusted for trade inventory), with discount brands +67% (Altria Q2 release above). A −9% to −11% Q3 would be a sharp deterioration. Watch it on 10-29; not used.
• [whiggity] MDLZ's hedged cocoa COGS still at peak levels in Q3/Q4 2026. Half right. On the Q2 call the CFO said "we would still see some phasing in on cocoa in Q3, which will reverse out in Q4", with EBIT up in both quarters, "more in Q4" (MDLZ Q2 2026 call transcript, 2026-07-28, via TickerTrends; secondary transcript). Q3 pressure is consistent; "peak in Q4" is contradicted. The MDLZ prepared-remarks PDF timed out.
• [whiggity] KHC: private-label down-trading, SNAP normalization, gas prices. No source given and none checked. KHC reports Wed 11-04, outside this week.
• [whiggity] Oral GLP-1 scanner drag on KHC/MDLZ/PEP. No source given; not checked.
• [whiggity] Section 232 aluminum tariffs on can economics (KO/PEP/bottlers). No source given; not checked. PEP's Q3 had $178M of tariff refunds, a 4-pt core OP tailwind that is fading (PEP remarks above), so tariffs matter. The aluminum-specific claim is open.
• [whiggity] MO trades as a bond proxy, so a hot PCE/10y can override an EPS beat by the open on 10-29. Plausible, given the 7:00 release and 8:30 GDP/PCE (verified). The "override" is untested; it is S-057.
• [whiggity] KO release at ~6:55 AM ET. KO only says "before the NYSE opens" (item 2).

Quant questions filed in strategy.md (all NEEDS DATA)

• S-055 (Q1): day-2 mean reversion of staples earnings reactions in high-oil regimes (Brent/WTI above its 1-year 90th percentile).
• S-056 (Q2): PM-follows-KO after controlling for DXY, 10y and XLP. Quick look done (item 10): 74% → 53%, and DXY is not the driver. Next steps: intraday timing and a larger peer set.
• S-057 (Q3, extends S-051): on earnings days that coincide with an 8:30 tier-1 macro print, split the premarket gap into pre-8:30 vs post-8:30 windows and test whether the macro surprise dominates MO's opening gap. First live case: MO Thu 10-29.
• S-058 (Q4): does the 60-day change in US retail gasoline prices predict the PM-minus-MO spread in earnings week?

Did the consult change the verdict?

No. "Not actionable today" stands, and nothing whiggity raised changes Monday's gap risk on a TRAI candidate:

• TRAI is flat by 3:55 PM. Today's fades (MO/PEP/TGT) are closed and carry no weekend exposure.
• The one new calendar fact is KOF moving to Mon 10-26 BMO. KOF is out of universe, so it is a Monday context read for KO's Tuesday print, not a trade.
• MO consensus is ~$1.50–1.51, not $1.42–1.45. The bar for Thursday is higher than whiggity implied. This does not change §6.3: do not fade MO's opening move on 10-29.
• The PM-follows-KO quick test shows the link is mostly the XLP factor, which weakens the KO-specific part of S-052. On Tue 10-27, treat PM as sector beta, not as a KO read-through. Tuesday's flag stays because PM still moves with XLP.

10. Unknowns / open questions

• KO and MO consensus figures come from secondary previews (MarketBeat, Yahoo/24/7, Hudson Labs, Zacks, S&P via moomoo). They agree with each other but were not checked against LSEG/FactSet.
• KHC 11-04 BMO is TRAI's date; not checked against KHC IR this hour.
• CL is not in the TRAI universe; CL→PG print-day linkage not computed.
• Peer-reaction stats are close-close on the reporter's reaction day; no intraday decomposition (e.g., vs 8:30 macro on 10-29).
• Live scoring uses full open→close; TRAI's rule bars entries before 10:00 ET, so actual results differ.

_Screening/research, not investment advice._