TRAI quant day-trading research

← Research · knowledge/reports/2026-10-23-1100-telecom-spacex-energy.md

2026-10-23 11:00 ET: Telecom carriers hit by SpaceX's tower talks, oil services jump on SLB and $116 Brent

Slot: Fri 2026-10-23, 11 AM ET (third hourly report). It moves away from the 9 AM semis report and the 10 AM software/consumer report. Status: FINAL (written ~11:05–11:30 ET). Includes a second-researcher consult with whiggity (Monica, new chat, ~11:10 ET); §8 lists which of its points were verified and which remain open. Live prices refreshed via Yahoo at ~11:15 ET where noted. Screening/research, not investment advice.

TRAI data used (pulled 2026-10-23 ~15:05 UTC): /api/latest is still the premarket run 2026-10-22_premarket_2026-10-23T12-53-19-641Z (8:53 ET; daily stats are for the Oct 22 session; no intraday run yet). Also used: /api/events (asOf 2026-10-23), /api/correlations (60-session window, asOf 2026-10-22), /api/regime, /api/macro, /api/scorecard, all under https://rva.guru/trai. Prints marked "Yahoo 11:06" come from the public Yahoo chart API (5-minute bars, last bar ~15:06 UTC; unofficial, may be delayed). TRAI uses the same feed.

1. Focus summary and why

  • Why this focus: at 11:06 ET the biggest movers in TRAI's 156-name universe were (a) the telecom cluster: T −5.74%, TMUS −4.47%, VZ −1.98%, CMCSA −1.14%; (b) oil services and E&Ps: SLB +5.76%, HAL +4.85%, COP +4.3%, DVN +4.3%, EOG +3.4%; and (c) the security-software cluster: PANW +6.3%, NET +6.1%, CRWD +5.3%, ZS +4.0% (Yahoo 11:06). Earlier reports covered none of these. VZ also reports Mon Oct 26 BMO and is the only imminent name for the next session (TRAI /api/events).
  • Telecom: Bloomberg reported late Oct 22 that SpaceX is in talks to secure at least 80,000 US cell sites, including with Crown Castle. That would solidify a path to a fourth nationwide wireless carrier ("Starlink Mobile") (Bloomberg via Yahoo, 2026-10-22 5:31 PM ET). Carriers fell and tower owners rose: CCI +9.5%, AMT +4.4% at 11:06, both well off their opens of +15.5% and +10.6% (Yahoo 11:06). At ~11:15: CCI +10.2%, AMT +4.9%, SBAC +4.3% (opened +9.9%).
  • Energy: SLB beat and is up ~5.8% (it opened only +0.8%, then ran). Brent is at a ~6-month high above $116 on Middle East escalation risk (CNBC, 2026-10-23).
  • Security: a second leg of the PHP-flaw ("PHP2Shell") rally that began Oct 21–22. I have not verified any catalyst specific to today (see §5).
  • Macro since the 10 AM report: no new scheduled US releases in the 10–11 ET hour that I could find. The 10y is ~5.00% (Yahoo ^TNX, 14:51 UTC), up from a 4.95% prior close, extending the post-PMI/UMich reversal covered in the 10 AM report.

2. Telecom: T, VZ, TMUS (CMCSA as a sympathy name)

Businesses. AT&T (T), Verizon (VZ) and T-Mobile (TMUS) are the three national US wireless carriers. T and VZ also sell fiber and fixed-wireless home internet. CMCSA (Comcast) is cable broadband plus an MVNO wireless business. Crown Castle (CCI), American Tower (AMT) and private Vertical Bridge lease tower space to carriers. CCI and AMT are not in TRAI's universe.

Catalyst (Oct 22 AMC, news, not earnings):

  • SpaceX "aims to secure a minimum of 80,000 cell sites across the US" and is talking with at least one major tower company, including Crown Castle. CCI CEO Hillabrant, asked on the Oct 21 earnings call about SpaceX/Amazon leases, said "You can assume that we're talking to all parties." Musk on SpaceX's call this week: satellite launches plus spectrum "will pave the way for Starlink Mobile to become a major mobile carrier in the US" (Bloomberg via Yahoo, 2026-10-22).
  • Scale: 80,000 sites is about half of the 158,500 purpose-built US towers. CCI has ~40,000 US towers, AMT ~42,000, Vertical Bridge ~18,000. TD Cowen estimates a nationwide terrestrial network would cost more than $80B and take many years (Reuters via Investing.com, 2026-10-22). The threat is long-dated, but it is now a re-rating factor for carrier multiples.
  • Spectrum: SpaceX signed an Oct 8, 2026 purchase agreement for up to 14 MHz of paired nationwide 800 MHz low-band spectrum (pending FCC approval) (SpaceX FCC narrative, 2026-10-08; SpaceX updates page). This adds to the EchoStar S-band (AWS-4/H-block) purchase announced Sept 2025 (same SpaceX page).
  • Premarket: T −4.3% to $21.76, VZ −4.3% to $43.25, TMUS −5% to $150.82, CCI +12% (MarketWatch, 2026-10-23). At 9:53 ET the S&P communication services index was −1%, carriers were −3% to −4.7%, CCI +12.6%, AMT +6.9% (Reuters via Investing.com, updated 2026-10-23 10:18 ET).
  • Intraday path (Yahoo 11:06, % vs Oct 22 close):
OpenLowHighLast
T−4.40−6.02−2.33−5.74 (near the low)
TMUS−5.16−5.78−2.50−4.47
VZ−3.54−3.93−0.27−1.98 (recovered about half)
CMCSA−1.62−1.93+1.09−1.14
IYZ (telecom ETF)+0.08

Dispersion: T is the weakest and VZ the strongest. The towers have given back much of their open.

AT&T context (reported Oct 21 BMO): revenue $32.3B (+5.3% y/y), adjusted EPS $0.65 (vs $0.54 y/y), GAAP diluted EPS from continuing ops $0.13 including a non-cash impairment on the held-for-sale Mexico wireless business plus employee-separation charges. 379,000 postpaid phone net adds, postpaid phone churn 0.92%, 442,000 fiber net adds. 2026 adjusted EPS guidance now "in the $2.40 range" (AT&T 8-K Ex. 99.1, SEC EDGAR, 2026-10-21). The 8-K text confirms a $2.6B non-cash impairment recorded at Corporate and Other when AT&T applied held-for-sale accounting to the Mexico wireless business (sale agreed Oct 13, 2026). It is $0.41 of the $0.52/share of adjustments. Total debt was $159.6B and net debt $149.8B at Sept 30, both up q/q on the closing of the EchoStar spectrum transaction (same 8-K). AT&T says it is "increasing its outlook for 2026 adjusted EPS". The prior range of $2.25–2.35 comes from Ad-Hoc-News, 2026-10-23, not the 8-K. Investing.com adds softer-than-expected postpaid phone adds and target cuts (MS $25, JPM $31, KeyBanc $33, Barclays $21), with today's session low at $21.58 vs a 52-week low of $19.89 (Investing.com, 2026-10-23 09:52 ET). Goldman Sachs cut its PT to $27 from $30 (Buy kept) on Oct 21, citing customer-acquisition costs and weaker postpaid phone adds (TipRanks/The Fly; Investing.com). TRAI /api/events Oct 21 reaction: gap −6.53%, close-to-close −6.45%, EPS surprise +7.75%. T then bounced +3.79% on Oct 22 (TRAI /api/latest), so today's −5.7% gives back that bounce and more. It is ~1.6 ATR (ATR 3.50%). Correction vs the draft: this is not a second straight down day. A TD Cowen survey found 52% of AT&T customers would consider switching to Starlink Mobile, vs 30% at T-Mobile and 23% at Verizon (Rolling Out, 2026-10-09, secondary source). That fits T being the weakest carrier today, but it is only a fit, not a cause I can show.

Verizon: reports Mon Oct 26 BMO (TRAI gapRisk `imminent`).

  • Consensus: EPS $1.28 / revenue $34.77B per Investing.com, with EPS estimates down 0.69% over 60 days and a mean PT of $51.24 (Investing.com, 2026-10-23 09:44 ET). Zacks has $1.27 / $34.81B and an ESP of −0.68% (Zacks via Yahoo). The sources don't fully agree; treat the consensus as roughly $1.27–1.28.
  • Guidance to test, from Q2 (raised for the second straight quarter): 2026 mobility and broadband service revenue +2.5–3.0%, "approach 3.0%" in Q3 and ~4.0% in Q4. Adjusted EPS $4.99–5.04. FCF growth 9–10%. Postpaid phone net adds in the upper half of 750k–1.0M. Capex $16.0–16.5B. Buyback up to $4.5B (VZ 2Q26 8-K Ex. 99, SEC EDGAR; Verizon newsroom).
  • Expect questions on Starlink Mobile and the VZ/T/TMUS satellite JV signed in October (Investing.com, 2026-10-23).
  • TRAI history, last 4 prints: avg |gap| 3.12% (1.76 ATR), fill rate 25%, continuation 50%, avg day-1 move in the gap direction +2.18%. All 4 prints gapped up: +4.09%, +5.53%, +2.26%, +0.59%, on EPS surprises of +1.5% to +5.8% (TRAI /api/events).

T-Mobile: reports Wed Oct 28 AMC (`elevated`). Last 4: avg |gap| 4.13% (1.50 ATR), fill rate 25%, continuation 75%. Jul 23 print: gap −7.96% and close-to-close −10.75% despite a +14.9% EPS surprise (TRAI /api/events). Today it is −4.5%, about 1.2 ATR (ATR 3.72%).

Sector linkages (TRAI /api/correlations, 60d):

  • Cluster 10 = {CMCSA, TMUS, T, VZ}, avg ρ 0.717. Pairs: T–VZ 0.90, VZ–TMUS 0.88, T–TMUS 0.84. CMCSA is 0.56 to each of the three carriers.
  • Telecom is a defensive/anti-semis factor. Its next-highest links are PEP/KHC/KO/NFLX (0.35–0.48), and it is about −0.38 to −0.42 vs ASML/AMAT/MU/ANET/DDOG. Beta60: T −0.33, VZ −0.41, TMUS −0.66 (TRAI /api/latest).
  • So today's carrier drop is idiosyncratic to the cluster. It is not a risk-off move: SMH is +1.3% and SPY −0.2% at 11:06.
  • Read-across: towers move inversely to carriers on this story. CMCSA's wireless/MVNO business takes partial sympathy.

Typical reaction: I have no TRAI history for news-driven (non-earnings) competitive shocks. Today's intraday pattern so far has VZ retracing about half and T not retracing at all. That split may track idiosyncratic baggage (T's post-earnings debt and impairment worries) rather than the SpaceX news itself. Interpretation, unsourced.

3. Energy: SLB beat plus $116 Brent; oil services lead E&Ps

Businesses. SLB and HAL (Halliburton) are oilfield services: drilling, completions, production systems and digital. SLB also has a growing Data Center Solutions unit. COP, DVN, EOG and OXY are E&P producers whose cash flows are leveraged to crude prices. XOM and CVX are integrated majors.

SLB Q3 (reported today BMO) (SLB IR, 2026-10-23):

  • Results: revenue $9.449B (+5% q/q, +6% y/y). GAAP EPS $0.49. EPS ex-charges $0.64 (+16% q/q, −7% y/y). Adjusted EBITDA $2.07B, margin 21.9% (+71 bp q/q). FCF $828M. Q3 buybacks $656M.
  • Regions: Middle East & Asia revenue $2.81B (+9% q/q). Middle East +13% q/q "despite late-quarter disruptions", with renewed disruptions in some countries in the final weeks. North America $2.40B (+7% q/q, +24% y/y).
  • Data Center Solutions revenue $230M (+23% q/q, +76% y/y). It includes a Meta delivery-partner facility in Canada opening 1H27.
  • CEO: "a very compelling outlook for 2027, with further upside as Middle East activity recovers."
  • FY26 capital investment is still ~$2.5B.
  • Consensus was ~$0.62 / $9.27B (Investing.com, 2026-10-23).
  • Q4 guide (call commentary, not in the press release): the press release gives no Q4 revenue figure; it only keeps FY26 capital investment at ~$2.5B. On the call, CEO Le Peuch said: "our fourth quarter outlook assumes Middle East revenue remains flat sequentially at approximately $1.9 billion … we expect overall fourth quarter revenue to reach approximately $10 billion, representing an increase of 5%-7% sequentially." He also guided adjusted EBITDA margin up 100–150 bp, Core up low single digits (Production Systems up high single digits; Well Construction and Reservoir Performance flat) and Digital up on year-end exploration sales. For 2027 he sees OFS market growth in the mid-single digits if the Middle East holds at 2H26 levels and the conflict "doesn't materially escalate". CFO: FY26 buybacks ~$2.4B (Investing.com transcript, published 2026-10-23 10:41 ET; Investing.com labels it AI-assisted and editor-reviewed). The official prepared-remarks PDF is linked from the SLB IR release page; I did not parse the PDF. "Middle East" (~$1.9B) is a subset of the reported "Middle East & Asia" area ($2.81B; the Middle East is 67% of it).

Price action:

  • SLB opened +0.83%, then ran to +6.01% high and was +5.76% at 11:06. HAL is +4.85% (Yahoo 11:06).
  • TRAI's 8:53 ET indicated gap was +2.00%. History: SLB fills its earnings gap 75% of the time (avg |gap| 2.21%). Today the gap was small and the move was intraday continuation, not a fill.
  • SLB ATR 2.85% means today's move is about 2 ATR. SLB–HAL ρ 0.80 (cluster 15). SLB's next-highest links are DVN 0.43, CVX 0.42, OXY 0.41 (TRAI /api/correlations).
  • HAL has no print in the window (next 2027-01-20, estimated). Its move is sympathy plus oil.

Oil:

  • Brent futures +2.6% to $116.13, the highest since Apr 30. WTI ~$102.56. Traders expect the US to resume strikes on Iran in the coming weeks, and Houthis keep targeting Saudi Arabia (CNBC, 2026-10-23).
  • Aramco ordered evacuation of its Dhahran HQ on Oct 22, and an oil spill was reported off the UAE (Reuters via Investing.com, 2026-10-23).
  • Brent was below $97 at the start of October. AAA gasoline is $4.29/gal vs $3.07 a year ago (AP via TT News, updated 2026-10-23 10:32 ET).
  • Brent $116 vs $111: resolved, it is a contract-month difference. Yahoo's BZ=F has already rolled to the January 2027 contract (BZF27.NYM quoted identically at $111.27). The December 2026 contract (BZZ26.NYM) was $116.01–116.07, +2.6% (Yahoo, ~11:05–11:15 ET), which matches CNBC's $116.13. ICE Brent December 2026 last trades Fri Oct 30, 2026 (futuresclock ICE Brent calendar; ICE expiry page). The Dec–Jan spread of ~$4.7/bbl (~4%) means steep backwardation: the prompt barrel carries a large premium. WTI (CL=F = Dec 26) is $102.10 and is not the source of the gap. (Verified from whiggity hypothesis #2, the front-month vs continuous roll; its alternative, "Yahoo showing WTI", is wrong.)

Linkages:

  • Energy is TRAI's anti-tech factor. /api/macro 20-day sector leaders: XLE +8.4% vs SPY, while SMH is −12.4%. Majors' beta60 is negative (XOM −1.07, COP −1.02; TRAI /api/latest).
  • Higher oil feeds the rates and inflation story: the 10y is back at ~5.00% after the hot PMI-price/UMich prints covered at 10 AM. That pressures the duration-sensitive housing/consumer names (HD −2.7%, LOW −2.4%, TGT −2.9% at 11:06).
  • Upcoming energy catalysts: PSX Oct 28 BMO (avg |gap| 1.0%); XOM and CVX Oct 30 BMO (avg |gap| 1.26% and 1.20%, ~0.6 ATR); COP/EOG/DVN Nov 5 (TRAI /api/events). The majors' earnings gaps are small. Their XOM fill rate is 75% with 0% continuation; CVX continuation is 100% (n = 4 each).

4. Typical post-catalyst reaction (TRAI /api/events)

  • Universe aggregate (n = 622): median |gap| 3.76%, fill rate 38.7%, continuation 49.2%, avg day-1 move in the gap direction −0.16%. For big gaps (median 8.75%), fill rate is 2.3%.
  • VZ (Monday): 4/4 gap-ups, avg day-1 in the gap direction +2.18%, fill rate 25%. TMUS: continuation 75%, but its largest recent gap (−7.96%) came on a beat. T: continuation 25%, fill rate 25%.
  • SLB: fill rate 75%, continuation 50% (n = 4). Today broke that pattern (small gap, strong continuation).
  • n = 4 everywhere, so treat all of these as anecdotes, not estimates.

5. Secondary: security-software second leg (PANW/NET/CRWD/ZS)

  • Background (Oct 21–22):
  • PHP maintainers disclosed two critical flaws, which Wiz, OpenAI and Quarkslab helped find (TipRanks, 2026-10-21/22). NSA and FBI urged immediate fixes or taking systems offline (CNBC, 2026-10-22).
  • Exploit attempts reached "millions" per Cloudflare's CEO, and Amazon also saw widespread exploitation (Reuters via Investing.com, 2026-10-22).
  • MS's Meta Marshall: "see tactical longs in the group." Overweight PANW, $410 PT (CNBC, same article).
  • Today: the move is +4% to +6.3% (Yahoo 11:06; HACK +2.8%). All four opened roughly flat (−0.2% to +1.0%) and rallied after 10:00. I found no today-specific catalyst. Search results conflated Oct 21–22 coverage with today. Unknown.
  • TRAI context:
  • Oct 22 was mixed: PANW +2.21, ZS +2.78, NET −1.18, CRWD −0.58 (TRAI /api/latest). None of the four was in a bucket.
  • Cluster 5 = {PLTR, DDOG, NET, ZS, PANW, CRWD}, avg ρ 0.633. PANW–CRWD 0.91.
  • The cluster carries high beta (CRWD 2.52, NET 2.64), so with SMH +1.3% part of today's move is beta.
  • Next prints: NET Nov 5 AMC; PANW ~Nov 19 (est.); ZS ~Nov 24 (est.); CRWD Dec 1 (TRAI /api/events). Nothing in the 5-day window.

6. ACTIONABLE for today (Fri 2026-10-23): screening notes, not advice

Prices are Yahoo ~11:15–11:18 ET (% vs the Oct 22 close). Bucket membership comes from TRAI /api/latest (8:53 ET premarket run; still the latest run at 11:16 ET).

  1. No TRAI bucket names are directly hit by this hour's catalysts. T/VZ/TMUS/CMCSA, SLB/HAL and PANW/NET/CRWD/ZS are absent from all three buckets (long_momentum: BX, SPOT, ADBE, HON, NOW, IBM; 12 short_momentum names; 17 mean_reversion names). There has been no intraday run yet.
  2. Telecom intraday shorts would land in TRAI's worst cell. T is −5.7% (~1.6 ATR) and TMUS −5.0% (~1.35 ATR), both near their lows. If a rescan flags either as short_momentum, the replay for short momentum is weak: in VIX-transitional regimes n = 329, mean −0.51%/trade, t −3.66 (TRAI /api/scorecard; 09:00 report §8). The 2-ATR target would also need T at roughly −12.7% on the day. T/VZ/TMUS form one cluster (ρ 0.84–0.90), so at most one name per cluster.
  3. T−VZ spread is a ~3σ day (quick check, mine). T−VZ log-return spread is −3.63 pp, z −3.17 vs the prior 252 sessions (Yahoo daily, sd 1.12 pp). In the last year there were 6 prior |z| > 3 days, almost all on T or VZ earnings dates. On the 4 days when T lagged, the next day's spread was +1.31, +0.14, +1.80 and +0.02 pp (T caught up slightly each time). n = 4, all event-driven, so this is an anecdote, not a signal (proposal S-021).
  4. VZ reports Mon Oct 26 BMO (`imminent`). TRAI is flat by 3:55 PM, so there is no direct exposure. Monday's premarket run will see VZ after a SpaceX drawdown and going into a print that gapped up 4 of the last 4 times. Read-through then hits T/TMUS (TMUS reports Oct 28 AMC).
  5. Towers are an extreme residual move, not a TRAI name. CCI +10.2% is a +5.0σ residual vs SPY and the Δ10y (252-day OLS; AMT +3.1σ, SBAC +2.2σ). All three gave back about half their opening gains. They are not in TRAI's universe, so this is context only (proposal S-019, quant test S-020).
  6. Energy services are ~2.4 ATR extended on an earnings day. SLB is +6.8% (ATR 2.85%, opened +0.83%) and HAL +5.2%. SLB's own Q4 guide assumes "no further escalation" in the Middle East. Oil is headline-driven into the weekend, and Dec Brent expires Oct 30 with ~$4.7 of Dec–Jan backwardation. Chasing here would mean the 2-ATR target is already used up. Not a TRAI signal.
  7. The defensive mean-reversion shorts are already moving the TRAI way, before any rescan. PEP is −2.4% (~1.1 ATR), MO −1.6% (~0.8 ATR), JNJ −1.7% (~0.7 ATR); HD −2.5%, LOW −2.3%, TGT −2.9% (~1.0–1.1 ATR). Telecom (ρ 0.3–0.45 to PEP/MO/KO) is falling with them, and the 10y is at 4.99%. As in S-013, much of the move came before 10:00, so the reward left on a 2-ATR target is shrinking.
  8. Factor offset. Semis are up (SMH +1.1%, AMD +2.2%, MU +3.2%) while telecom and staples are down. The mean-reversion longs (AMD/NVDA/MU/ASML/INTC) and the defensive shorts are currently winning together. That is one factor bet (anti-defensive / pro-semis), not diversification (S-014, S-018).

7. Unknowns / to verify

  • Today-specific catalyst for the second leg of the security rally. None found.
  • Bloomberg report: no timeline or deal terms. 80,000 sites is a minimum target, talks are private, and no deal is confirmed.
  • VZ consensus differs by source ($1.27 vs $1.28 EPS; $34.77B vs $34.81B revenue).
  • SLB prepared-remarks PDF not parsed. The Q4 ~$10B figure rests on the Investing.com transcript.
  • AT&T's prior FY26 adj. EPS range ($2.25–2.35) is from Ad-Hoc-News only.

8. Second-researcher consult: whiggity (Monica), new chat, ~11:10 ET

Kieran's Monica research bot whiggity was consulted in a new chat at ~11:10 ET with this report's draft summary. Its reply is treated as a second researcher's input, not as a source. Below, items are marked [whiggity] and sorted by what I could verify against public sources (accessed 2026-10-23 ~11:12–11:20 ET).

Verified (kept, with citations)

  1. [whiggity] AT&T $2.6B impairment: verified (primary source). It is non-cash and tied to the held-for-sale Mexico wireless business, recorded at Corporate and Other, after the Oct 13, 2026 sale agreement (to Grupo Televisa and Fintech Holdings). Q3 revenue was $32.3B and adj. EPS $0.65, and AT&T is "increasing its outlook for 2026 adjusted EPS" to "the $2.40 range" (AT&T 8-K Ex. 99.1). whiggity's cited secondary sources agree (Ad-Hoc-News; El Fondo, which puts consensus at $0.61 EPS and $32.32B revenue). The Kalkine link timed out and was not read.
  2. [whiggity] Goldman PT cut: verified (secondary sources). $30 → $27, Buy kept, Oct 21 (TipRanks/The Fly; Investing.com; Ad-Hoc-News citing Markets Insider). The original Goldman note was not seen.
  3. [whiggity] Brent $111 vs $116: verified as a contract roll. Yahoo BZ=F = Jan-27 ($111.27); Dec-26 (BZZ26.NYM) = $116.0, matching CNBC. Dec-26 last trading day is Oct 30, 2026 (futuresclock). Steep backwardation (~$4.7 Dec–Jan) is confirmed from the two Yahoo quotes. whiggity's alternative ("Yahoo shows WTI") is wrong: WTI is $102.
  4. [whiggity] SLB Q4 ~$10B: verified as call commentary, as whiggity's caveat required. The figure is not in the press release (SLB IR). It is in the CEO's call remarks: Q4 ~$10B, +5–7% q/q, assuming Middle East flat at ~$1.9B and no further escalation (Investing.com transcript).
  5. [whiggity] Crown Castle ~40,000 US towers: verified. AMT has ~42,000 US sites and Vertical Bridge ~18,000 (Reuters via Investing.com, 2026-10-22). Arithmetic: 80,000 sites is more than CCI + Vertical Bridge (58,000), so SpaceX needs AMT and/or SBAC sites, or many non-tower sites. The SBAC count was not verified.
  6. [whiggity] TMUS is SpaceX's direct-to-cell partner: verified. T-Satellite runs on Starlink, with exclusivity through at least end-2026 (Light Reading, 2026-09-10). T-Mobile removed Starlink branding from the T-Satellite page, but the partnership has not been ended (Rolling Out, 2026-10-09). "Partner becomes competitor" as the reason TMUS is down is whiggity's interpretation and is not sourced. T is down more than TMUS.
  7. [whiggity, counter-evidence] SpaceX's spectrum and network gap: partly verified. TD Cowen says SpaceX "would need several times more spectrum" and >$80B to build a network. AT&T/VZ/TMUS refuse it a wholesale MVNO deal (Reuters analysis, 2026-10-16; Light Reading). Correction: "lacks terrestrial mid-band spectrum" is too strong. SpaceX holds EchoStar's AWS-4/H-block (~2 GHz) plus the pending 800 MHz purchase; the constraint is quantity, not band. The DISH/EchoStar fourth-carrier history is background, consistent with EchoStar selling spectrum to AT&T and SpaceX (AT&T 8-K, SpaceX updates page).

Open questions (not verified; not used as facts)

  • [whiggity] CCI's US-only exposure explains why CCI beat AMT. Plausible (AMT is more international), but no source. Note that the residual vs SPY/rates also ranks CCI > AMT > SBAC.
  • [whiggity] FWA offload rather than a mobile price war. No source.
  • [whiggity] SLB Gulf/Red Sea war-risk insurance exposure. Not checked. SLB's own call flags renewed Middle East disruptions and conditions its guide on no further escalation (verified).
  • [whiggity] HAL is NA-land leveraged and overextended in sympathy. No source checked. HAL has no print in the window (next 2027-01-20, estimated; TRAI /api/events).
  • [whiggity] A vulnerability rally rarely turns into same-quarter ARR. No source.
  • [whiggity] Related names SBAC, ASTS, GSAT, CHTR, CMCSA, BKR, WFRD, AKAM, FSLY. Quotes at ~11:15 ET (Yahoo): SBAC +4.3%, ASTS +0.5%, GSAT −0.1%, CHTR −3.5%, CMCSA −1.2%, BKR +3.7%, WFRD +7.5%, AKAM +1.6%, FSLY +5.4%. The claim that cable MVNOs ride Verizon's network was not re-verified this hour. Of these, only CMCSA is in TRAI's universe.
  • [whiggity] Quant questions (a)–(d): see the quick checks below and proposals S-020 to S-022.

Quick checks run on whiggity's quant questions (mine; Yahoo daily bars, not TRAI's engine)

  • (a) Tower residual z (252-day OLS on SPY and Δ10y): CCI +10.5% residual, z +5.0; AMT +5.3%, z +3.1; SBAC +4.8%, z +2.2. The rates betas are negative, as expected for REITs (about −5% to −11% per 1 pp of 10y). Today's 10y move (+4 bp) explains almost none of the move.
  • (b) T−VZ spread: z −3.17 today (see §6.3). Prior |z| > 3 days: n = 6 in 252 sessions, all near earnings. Too few to test before VZ prints Monday.
  • (c) SLB "quiet open → strong 11 AM" earnings days and (d) cyber basket holding +4% into a Friday close with the 10y ≥ 5%: not tested. They need intraday history (proposal S-022).

Proposed strategy notes (added to strategy.md as PROPOSALS)

  • S-018: news-driven shocks to defensive clusters. Exclude a cluster from defensive-hedge netting and from same-day fades when it drops > 1.5 ATR on non-earnings competitive news.
  • S-019: context symbols outside the universe. Add CCI/AMT/SBAC (and the SpaceX ticker) as non-tradable read-through symbols for telecom news.
  • S-020: residual z for context names (whiggity 8a).
  • S-021: pair-spread shocks before a peer's print (whiggity 8b).
  • S-022: intraday "quiet open, strong 11 AM" earnings continuation (whiggity 8c/8d).

Sources (all accessed 2026-10-23)