Trade Construction

Oracle Corporation [ORCL]

Oracle Corporation [ORCL] — Trade Construction

No position is recommended. This memo issues no verdict — it scores Criteria and hands the analysis to the book. What follows is the implementation study the book needs if it decides to act, with every instrument priced from a chain actually pulled.

Spot $119.95 (2026-07-28 close). Market cap $345.5bn.

1. What the analysis supports, and what it does not

	Reading
Valuation Criteria	PASS on the run-rate (+4.4pp), PASS WITH ARGUMENT on the 3-year CAGR (−5.7pp)
12-month target	$134, +11.9%
Downside case	$72–98, −40% to −18%, named cause: counterparty non-performance
Momentum Criteria	negative — 12-1 momentum −30.6%, 0.65x of the 200-day, 4.3% off the 52-week low
Reward / risk at the base and bear	+11.9% against −18% to −40% → 0.30x to 0.66x

The honest summary: the ownership test passes and the twelve-month risk/reward does not. That combination is what the Momentum Criteria exists to express — it governs when to enter a position the thesis already justifies, never whether. A −30.6% 12-1 momentum in a name 4.3% off its 52-week low is a falling knife, and the correct output is a staged entry conditioned on a dated event, not a full position today.

2. Volatility and the vehicle decision — chain pulled 2026-07-29

Realised volatility (252d): 65.8%. Beta 1.85.

ORCL 16-Jan-2027 calls (Alpaca options snapshots, live quotes and open interest):

Strike	Open interest	Bid	Ask	Spread	IV	Delta
110	368	25.72	27.11	5%	74.1%	0.652
120	1,739	21.38	22.65	6%	73.2%	0.585
130	3,554	17.69	19.20	8%	73.0%	0.521
140	3,294	14.54	16.06	10%	72.4%	0.460
150	11,618	12.16	13.20	8%	72.0%	0.403
160	6,066	9.69	11.04	13%	71.2%	0.350
170	4,414	8.54	9.31	9%	72.2%	0.310

The chain is investable. Open interest of 1,700–11,600 on round strikes with 5–10% spreads is a different world from the HCA failure the Liquidity Criteria was written for (18 contracts across an entire chain).

But the premium is not cheap and it is not expensive either: IV ≈ 72% against 65.8% realised, a ratio of 1.10. That is a normal, not a punitive, volatility risk premium. Long premium is permissible here — which is unusual and worth saying, because on most names in this book it has not been.

3. The structures, priced

(a) Equity, staged — the reference case

Buy in three tranches of one-third each: on any close above the 50-day; on the first quarterly print that shows FY2027 OCI revenue tracking to the published $32bn path; on any move to $100 or below (the downside-case zone) provided the RPO duration disclosure has not deteriorated. Invalidation: a close below $110 (a 4.3% break of the 52-week low of $115.00 plus a buffer) with the counterparty risk unresolved → the de-rating is discounting something the filings do not show, and the correct response is to stop, not to average.

(b) Jan-2027 $130/$170 call spread — the defined-risk expression

Buy 1 × $130 call at the ask 19.20; sell 1 × $170 call at the bid 8.54.
Net debit $10.66. Max value $40.00. Max profit $29.34. Risk/reward 2.75 : 1.
Breakeven $140.66, +17.3% from spot. Max payoff above $170, +41.7%.
Both legs liquid: OI 3,554 and 4,414; spreads 8% and 9%.
Why this and not the outright: it monetises the exact zone between the 12-month base ($134) and high ($156) case while capping the cost of being early, which the negative momentum makes likely. It requires more than the base case to pay, and that is stated as a drawback, not hidden.

(c) Jan-2027 $120/$150 call spread — the base-case expression

Buy $120 at 22.65; sell $150 at 12.16. Net debit $10.49. Max value $30.00.
Max profit $19.51. Risk/reward 1.86 : 1. Breakeven $130.49, +8.8% — below the $134 base target, which is the point.
OI 1,739 / 11,618. The $150 strike is the most liquid line on the board.

(d) Rejected, with reasons

Front-week or front-month premium: rejected. Nothing in the thesis resolves inside a month; the RPO disclosure is quarterly.
Outright Jan-2027 $130 call at 19.20: rejected relative to (b). It costs 80% more for the same directional view and the $170 strike sold in (b) sits above the 12-month high case of $156, so almost nothing of value is given up.
Put spreads / any short expression: rejected. The Valuation Criteria passes; shorting a name whose implied path is achievable, 63.5% below its high and 4.3% off its low, is shorting the de-rating that has already happened.
Selling puts to finance: rejected. The named downside case is −40% and it is a counterparty credit event, i.e. exactly the tail that short puts are worst against.

4. Sizing

Inverse-volatility sizing is the interim protection the Downside Criteria relies on. At 65.8% realised volatility ORCL is a HIGH volatility tier name notwithstanding its $345bn market cap — and that is the finding, not a technicality.

Conviction is MEDIUM: the mechanism is filed and dated, the first year of the published path was delivered, and the multiple is at the 1st–17th percentile of its own history — but the load-bearing input (single-counterparty RPO concentration) is not disclosed at all, and the momentum is against.

MEDIUM conviction × HIGH volatility, staged, with a 40% first tranche. Final size is the book's call under portfolio-book, not this memo's. No entry is appended to trade_recommendations.jsonl, because no position is recommended.

5. Liquidity Criteria — formal score

Test	Result
Equity liquidity for intended size and a plausible exit	PASS — mega-cap, $345bn
Options chain actually pulled before any structure proposed	PASS — pulled 2026-07-29, Jan-2027 expiry, 13 strikes
Open interest and quoted size at the specific strikes proposed	PASS — 1,739–11,618 OI, 5–13% spreads
IV versus realised at the proposed strikes	PASS — 1.10x, not punitive
Borrow / short interest	not applicable — long-only fork

6. Correlation note for the book

ORCL, INTC and DELL were underwritten as one batch and share an end-market, so the book must not treat three passes as three independent ideas. Correlation was not computed in this run — the 252-day pairwise matrix against existing book positions is a portfolio-book task and is flagged here rather than fabricated. What is certain from the price history: all three re-rated on the same AI-capex narrative within the same twelve months (ORCL +198% then −63%; INTC +347%; DELL +253%), so the prior should be high positive correlation, and the cluster should compete for one slot between them, not three.

Generated from each name's analysis.json. Figures are research output, not investment advice. Gaps are shown as “—” rather than filled: a missing input is recorded as indeterminate, never estimated.