Week of Aug 3, 2026

Strong bull signal: ROR +42%, Growth 60%, Seer score. AMD's Advancing AI 2026 event (7/23) raised the server-CPU TAM estimate to >$120B by 2030 and cited up to 2GW of Anthropic and 6GW of Meta compute commitments; Wedbush raised 2026-27 revenue/EPS estimates after the event. The stock has already pulled back 17% from its June 30 all-time-high close, so some enthusiasm has cooled — a call spread caps cost against a genuinely large historical move (the May 2026 print moved 25.9% vs. a 7.0% implied move, and AMD has blown through implied in 3 of the last 8 prints). Active CSP: $490 put exp 2026-08-21, low conviction — spot is currently well below $490, so that short put is already at risk. This bull spread runs the same direction (both bet AMD stabilizes or recovers), not against it, but it's worth watching both legs together.

• Risk: Implied-move estimates for this name ranged from 7.8% to 12.3% across sources this week — confirm the live number before pricing strikes; if it's toward the low end, a tighter spread captures more of the move for the same debit.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Textbook Tier 0 compression: Seer score, undervalued 50% vs Hist_Normal_PE, Miss 25% (<30%), ROR +49.2%. The stock is down 34% YTD despite 90% of 20 covering analysts rating it Buy, a Q2 guide of $1.9–1.95B revenue at 84–85% adjusted EBITDA margin, and the Axon 2 AI ad-engine rolling out in full. This is the valuation-plus-quality combination the compression pattern is built to catch. Active CSP: $350 put exp 2026-08-21, high conviction, no longer qualifies — this long call runs the same direction as that existing income position, not against it.

• Risk: Actual post-earnings moves have exceeded the ~12% implied move (some desks show up to 15.7%) in roughly half of the last 8 prints, in both directions — this is a real binary bet on a volatile name. Size stays at the Tier 0 ceiling, not above it.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Seer score (moderate), overvalued +112% vs Hist_Normal_PE, Miss 36% (high execution risk), bear score. Unlike ROK/HWM this one has real fresh bearish tape underneath the valuation signal: Michael Burry disclosed a new short position in July, Resource Industries segment profit fell 39% YoY with margin down 700bps, Power & Energy margin fell 170bps sequentially, and there was ~$87M of insider selling. The offset is a record $63B backlog (+79% YoY) on data-center power demand — that's the bull case that could produce a squeeze, so the short call needs to sit outside the ~6.1% implied move, not at it.

• Risk: The backlog number is genuinely strong and could dominate the headline if management leans on it. Miss rate (36%) cuts both ways — CAT has surprised to the upside before.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Tier 0 stale-norm pattern: overvalued +189% vs Hist_Normal_PE (a base depressed by the prior biosimilar-overhang years — not a meaningful ceiling today), but ROR +102%, Seer score, Growth 67.8%, Miss 18%. HSBC upgraded GILD to Buy on 7/6 citing HIV franchise outlook; the FDA accepted the sNDA for oral once-weekly lenacapavir; Trodelvy just got a first-line cancer label expansion. Headline GAAP EPS this quarter is expected to show a large loss (~-$7/share) from one-time IPR&D and acquisition charges — that's accounting noise, not core weakness; Biktarvy (the actual franchise) is still growing (~$3.6B, +7% YoY, >52% US share).

• Risk: The GAAP loss headline could spook algorithmic/retail reaction on print even though it's non-cash and expected — keep the long call ATM so a same-day overreaction doesn't wipe the position before it can recover.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Tier 0 stale-norm pattern: overvalued +73% vs Hist_Normal_PE, but Fore_Est_Annual_ROR +51%, Seer score, Growth 32.6%. Keytruda (including the newer Qlex formulation) is still growing double digits and just drew a price-target raise to $142 from BMO. Gardasil's China weakness is the well-known overhang — 2025 sales fell ~39% YoY there and shipments are paused — but that's already baked into FY2026 guidance, not a surprise. Implied move is low (~3.9%), consistent with a name the market expects to react quietly. Active CSP: $125 put exp 2026-08-21, low conviction — this bull spread runs with that position, not against it.

• Risk: Adjusted EPS this quarter is expected down ~36% YoY on the Gardasil drag. If China commentary worsens rather than just repeats the known story, the "quiet reaction" assumption breaks. Keep the short call inside 10% OTM rather than reaching for more premium.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Best quality name in the universe (Seer score, Miss 20%), but Fore_Est_Annual_ROR is only +8.4% — inside our neutral band, so this isn't a formula-driven Tier 0/1 pick. The reason it's here at all is news the reversion-based model can't see: the Army selected Palantir Foundry as the core data layer for its NGC2 modernization program, Palantir partnered with Nvidia on Foundry/Nemotron, and Palantir won a contested-solicitation protest against the DIA. DA Davidson upgraded to Buy ($175 PT) and the median 6-month analyst target (~$200) sits well above the $123 spot. Implied move is elevated (~15%), which argues for collecting premium via a credit spread rather than paying up for a directional call on an already-expensive multiple.

• Risk: The ~15% implied move is genuinely two-sided — PLTR's own history includes sharp post-earnings drawdowns on any guidance disappointment despite the news flow above. Keep size at Tier 3.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

Not in the local dataset — this is a pure web-research pick. Spotify is down 38% from its 2025 peak and 16% YTD, but Street sentiment hasn't followed: 34 of 40 analysts rate it Buy-or-higher with an average target of $597.60 (+23% from spot) and BofA at $685 (+41%). New AI-powered music/podcast features and continued price-hike cadence support the growth case. Implied move is elevated (~8.6–8.85%), and actual moves have exceeded implied in 5 of the last 8 quarters — real gap risk, which argues for a defined-risk credit spread over an outright long.

• Risk: SPOT has genuine two-sided history on earnings day; this is a "depressed vs. targets" thesis, not a clean technical setup. Confirm live implied move before sizing up.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.

ROR +57.7%, Growth 53.8%. NAND/HDD pricing-cycle strength is driving sharp upward FY26 estimate revisions (consensus EPS moved from roughly $9.04 to $11.89), reports of Kioxia merger discussions (7/9) could reshape the NAND landscape further, and Citi (PT $800) and Wells Fargo (PT $730) both raised targets in July. The dataset's overvaluation read (+419% vs Hist_Normal_PE) is a stale-trough artifact from the prior NAND downturn, not a meaningful ceiling — leaning on ROR, growth, and fresh analyst targets instead. Active CSP: $480 put exp 2026-08-21, low conviction — spot has traded through that level recently, so the existing short put is live risk; this spread runs the same direction.

• Risk: This is one of the most volatile names on the calendar (52-week range spans more than 10x) — use a defined-risk spread, not a naked long, and confirm live implied move before sizing.

From the written analysis. The weekly analysis made this call. The rule-based screen did not select this ticker and structure that week.

6 lessons were active that week. None applied to this play.