Oracle prices $638B of RPO, but the funding bridge blocks a clean short

Run timestamp: 2026-09-06 14:51 SGT Scope: U.S. market, short only Classification: No-trade screen, Reject for current execution Primary instrument: Oracle common stock, NYSE, unlevered only if every short-execution gate later passes

Summary: Oracle enters its September 10 fiscal Q1 FY2027 report with a market capitalization of about $457.36 billion and a stock that rose from $141.32 on September 1 to a $158.78 close on September 4. The investment case now leans on $638 billion of remaining performance obligations, or RPO, while the last filed year produced $31.98 billion of operating cash flow against $55.66 billion of capital expenditures and negative $23.69 billion of free cash flow.

The short disagreement is not that Oracle lacks demand. Its official release says Q4 cloud infrastructure revenue grew 93%, FY26 revenue grew 17%, and RPO rose 363% to $638 billion. The disagreement is whether a giant, partly customer-funded RPO number can be converted into profitable Oracle-owned capacity quickly enough to justify the current equity value while the company expects to raise about $40 billion of debt and equity in FY27, including a previously announced $20 billion at-the-market program.

The short is not executable. The regular-session close was $158.78 at 16:00 EDT on September 4, and the after-hours observation was $159.70 at 19:59 EDT. The latest public short-interest report shows 47.77 million shares short, 2.81% of public float, and 1.6 days to cover at the August 14 settlement. That is enough positioning to amplify a disappointment, but not enough to make squeeze risk small. Using $158.78 as a reference, the modeled top, base, and bottom prices are $125.00, $140.00, and $190.00 at probabilities of 30%, 45%, and 25%. The base decline is 11.83%, the adverse rise is 19.66%, and gross reward to adverse risk is 0.60:1 before costs. The Desk therefore rejects the short under the 10/5 hurdle and keeps entry.price null and execution.can_execute false.

Opportunity Ranking

The search used mechanism-specific queries rather than a generic earnings query:

RPO customer prepayment supplied GPU capex debt equity financing cash conversion earnings short setup

department store inventory real estate debt earnings short float September 10 2026

footwear ticker change inventory normalization debt-free cash high short float earnings catalyst

The current-week archive and September signal file were scanned before selection. Existing current-week names and mechanisms, including KOD's binary biotech cash clock, IONS's failed clinical endpoint, WFRD's redomestication bridge, KFY's cash repricing, USAR's resale shelf, SUNB's rental cycle, and the existing software and retail notes, were excluded. ORCL has no prior article or signal in the archive. M and SHOE are comparison candidates with distinct retail and inventory evidence.

Rank	Candidate	Direction	Mispricing	Evidence Freshness	Catalyst Window	Positioning	10/5 Status	Tradeability	Main Rejection Risk
1	Oracle (ORCL)	Short	The market can treat $638B of RPO as near-cash-like future value while FY26 free cash flow was negative $23.69B and FY27 still requires large debt and equity funding	Sep. 2 earnings-date release, June 10 FY26 financial release and tables, Sep. 4 tape	Q1 FY27 results on Sep. 10 after close	2.81% of public float short and 1.6 DTC as of Aug. 14; current borrow and dealer positioning unknown	Reject	NYSE large-cap with about 26.0M shares traded on Sep. 4, but current quote, borrow, and event exit path are unverified	Cloud growth, customer prepayments, and supplied hardware can make the RPO economics better than the headline cash burn suggests
2	Macy's (M)	Short	A Q2 earnings catalyst meets $2.4B debt and a middle-income consumer risk, but Q1 comparable sales, cash, and guidance improved	Aug. 19 earnings-date notice, Jun. 3 Q1 release, Sep. 4 tape	Q2 results on Sep. 10 before open	11.66% of public float short and 6.8 DTC as of Aug. 14	Reject	NYSE and about 3.53M shares traded on Sep. 4	$1.3B cash, $2.0B available borrowing capacity, and raised FY26 outlook
3	Shoe Station Group (SHOE)	Short	A Q2 earnings catalyst meets weak sales and margin pressure, but the company has no debt, $129.3M cash, and a high short float that can squeeze	Aug. 27 earnings notice, May 21 Q1 release, Sep. 4 tape	Q2 results on Sep. 10 before open	36.80% of public float short and 13.7 DTC as of Aug. 14	Reject	Nasdaq and about 1.02M shares traded on Sep. 4, with a recent ticker transition	Inventory reduction, debt-free liquidity, and improved banner trends can overwhelm the bear case

Selected research subject: ORCL.

Why this one now: It has the freshest confirmed catalyst, the largest documented funding bridge, and an unusually clean distinction between contracted demand, customer-funded hardware, Oracle-funded capex, and common-shareholder claims. It is also liquid enough to study without pretending that liquidity solves the valuation problem.

What should surprise the reader: The $638 billion RPO headline is not the bear's weakest point. Oracle explicitly says $75 billion of the large AI contracts were prepaid by customers or used customer-supplied hardware. The harder question is the undisclosed split between those two states. A prepayment can improve cash timing. Customer-supplied hardware can reduce Oracle's capital burden but may also change asset ownership, depreciation, margin, and revenue economics. RPO is a contract state, not a cash-flow statement.

No-trade conclusion: ORCL is the best current short research subject by evidence freshness, funding tension, and tradeability, not the best executable short. The base price-only decline clears 10%, but the modeled adverse rise is nearly 20% and the gross ratio is below 1:1.

Why This Is the Best Opportunity Right Now

The strongest counterparty case is concrete. Oracle's June 10 release reported FY26 revenue of $67.4 billion, cloud revenue of $34.0 billion, cloud infrastructure growth of 77%, operating cash flow of $32.0 billion, and Q4 RPO of $638 billion. It guides Q1 FY27 total revenue growth of 27% to 29% and total cloud revenue growth of 58% to 64% in U.S. dollars. It also says the prepaid and customer-supplied hardware portions of its large AI contracts total $75 billion and substantially reduce the capital Oracle must raise. Oracle FY26 results

That is not a weak business. The short case begins one layer later. The same official financial tables show FY26 operating cash flow of $31.977 billion and capital expenditures of $55.663 billion, leaving free cash flow of negative $23.686 billion. The balance sheet shows $31.289 billion of cash, $605 million of marketable securities, $7.199 billion of current notes and other borrowings, $122.342 billion of non-current notes and other borrowings, and $26.648 billion of operating lease liabilities at May 31. Oracle FY26 financial tables

Oracle funded that gap in FY26. The release says it raised $43 billion of debt financing and $5 billion of equity financing. For FY27, it expects to raise approximately $40 billion through debt and equity, including a previously announced $20 billion ATM equity issuance. The company says it does not expect to issue additional debt in calendar 2026, but that does not eliminate equity issuance, commercial paper, short-term capex financing, or the funding requirement itself. Oracle FY26 results

The September 10 report can close the gap in either direction. A cloud beat with RPO growth, stable margins, visible customer prepayments, and no deterioration in financing terms can validate the market's enthusiasm. A revenue beat with weaker cash conversion, a higher capital burden, lower-quality RPO composition, or a larger equity requirement can show that the market capitalized a contract headline before the common-equity funding cost.

Macy's has more obvious short interest but a stronger near-term cash and sales countercase. SHOE has even more short interest, but its debt-free balance sheet and inventory plan make the squeeze path dominant. ORCL wins because the current market is paying for a documented growth story while the company has already disclosed the exact cash and funding mechanics that can falsify it.

Why This Can Move More Than 5% Soon

The stock rose 12.35% from $141.32 on September 1 to $158.78 on September 4, including a 3.08% gain on 26.046 million shares on the last session. The move is not itself a short thesis, but it increases the distance between the current price and the last pre-rally reference. The next confirmed event is Q1 FY27 earnings after the close on September 10. Oracle price history, Oracle Q1 FY27 earnings date

The event can move more than 5% through four linked variables:

Cloud infrastructure growth must land inside or above the 58% to 64% guide range while the prior-year base becomes much larger.
RPO must convert into recognized revenue, usable cash, and acceptable margins. The quarter can show bookings without resolving the long-term return on the capacity.
The $75 billion of prepaid and customer-supplied hardware must be understood as a composition, not a single cash offset. Oracle does not disclose the split in the release.
The market must price the next funding step. A $20 billion ATM is capacity and a disclosed plan, not the same as completed issuance, but its existence changes the common-share denominator risk.

10/5 Asymmetry Gate

The map is measured from the completed regular-session close. The after-hours observation is kept separate and is not an entry.

Measure	Reference-only result	Interpretation
Reference price	$158.78	StockAnalysis regular-session close at 2026-09-04 16:00 EDT
Latest after-hours context	$159.70	StockAnalysis observation at 2026-09-04 19:59 EDT; non-executable
Top-case target	$125.00	21.27% favorable underlying decline
Base-case target	$140.00	11.83% favorable underlying decline
Bottom-case target	$190.00	19.66% adverse underlying rise
Gross reward to adverse risk	0.60:1	($158.78 - $140.00) / ($190.00 - $158.78), below the required 2:1
Probability-weighted target	$148.00	30% top, 45% base, 25% bottom; price-only estimate
Price-only expected decline	6.79%	($158.78 - $148.00) / $158.78 before costs
Measurement basis	Reference-only	`entry.price` is null
Status	Reject / no-trade screen	Adverse path and financing uncertainty are decisive

The base case clears the favorable 10% screen, but the adverse case is almost 20% and the gross ratio is 0.60:1 before costs. The positive price-only expected decline is not net expected value. A responsible net EV cannot be computed because the entry, borrow, spread, slippage, funding, event path, RPO composition, and post-report cash state are not verified.

What Should Surprise the Reader

Oracle generated $31.98 billion of FY26 operating cash flow and still reported negative $23.69 billion of free cash flow because capex reached $55.66 billion. Profit and operating cash do not by themselves finance the AI buildout.
The $638 billion RPO figure is not all Oracle-funded demand. Oracle says $75 billion of large AI contracts involve customer prepayment or customer-supplied hardware, but it does not disclose the split between those two states in the release.
The funding risk is not a simple solvency call. Oracle held $31.29 billion of cash and $605 million of marketable securities at May 31, against $129.54 billion of notes and other borrowings plus $26.65 billion of operating lease liabilities. The question is return on funded capacity and the cost of the next dollar, not whether the company can pay a bill tomorrow.
The $20 billion ATM is financing capacity, not completed issuance. The correct denominator test is the actual filing of shares, proceeds, and use of cash.
Short interest is not the main asymmetry. Only 2.81% of public float is reported short, so the adverse path can come from fundamental re-rating and momentum, not only a squeeze.

The Setup

Oracle's FY26 results are the most recent completed operating state. Q4 total revenue was $19.184 billion, Q4 cloud revenue was $9.913 billion, and cloud infrastructure revenue was $5.8 billion. FY26 total revenue was $67.357 billion, cloud revenue was $33.989 billion, and software revenue was $24.541 billion, down 1% in U.S. dollars. The company is shifting mix from legacy software toward infrastructure and applications in the cloud. Oracle FY26 financial tables

The RPO jump was $85 billion in Q4, from $553 billion to $638 billion, and up 363% year over year. The release states that most of the Q3 and Q4 increase came from large AI contracts where customers prepaid Oracle for GPUs or bought and supplied the GPUs. The company says those portions total $75 billion and reduce the capital it must raise. That is a positive funding fact, but it is not the same as recognized revenue, operating margin, or owner return.

The audited financial table shows the associated capital state. Oracle's property, plant, and equipment net rose to $99.957 billion from $43.522 billion one year earlier. Notes and other borrowings rose to $129.541 billion from $92.568 billion. Cash and marketable securities rose to $31.894 billion from $11.203 billion, but the increase was accompanied by $40.284 billion of net financing cash inflow. This is a funded expansion, not a self-financing one.

The common-share claim also changed. FY26 diluted weighted-average common shares were 2.914 billion versus 2.866 billion in FY25. Oracle issued $4.954 billion net of mandatory convertible preferred stock during FY26, alongside $1.449 billion of common-stock proceeds. The preferred security and the proposed ATM are not interchangeable, but both matter to the eventual common-equity denominator and claim priority. Actual future issuance remains unknown.

The next report is not a remote catalyst. Oracle announced on September 2 that Q1 FY27 results will be released after the close on September 10, with a 4:00 p.m. Central conference call. The official FY26 release gives the guide: total revenue growth 27% to 29%, cloud revenue growth 58% to 64%, and non-GAAP EPS of $1.72 to $1.76 in U.S. dollars. Oracle Q1 FY27 earnings date, Oracle FY26 guidance

The Market Price

Field	Observation	Timestamp or date	Source and limitation
Regular-session close	$158.78	2026-09-04 16:00 EDT	StockAnalysis ORCL history; completed market print
Regular-session change	+$4.74, +3.08%	Same session	Same source
Regular-session range and volume	$156.71 to $159.70; 26,045,939 shares	Same session	Same source; volume does not prove exit capacity
Extended-hours observation	$159.70, +0.58% from close	2026-09-04 19:59 EDT	StockAnalysis ORCL statistics; non-executable context
Market capitalization	$457.36B	Checked after the Sep. 4 close	StockAnalysis ORCL statistics; provider field
Enterprise value	$592.90B	Checked after the Sep. 4 close	Same source; provider methodology, not a substitute for the filed claim ledger
Shares outstanding	2.880471B	2026-06-12	Oracle FY26 Form 10-K; later changes unknown
Public short interest	47.77M shares, 2.81% of float, 1.6 DTC	2026-08-14 settlement	MarketBeat ORCL short interest; stale to the September event

The price has moved sharply into the catalyst. A regular close is available, but the current bid, ask, spread, depth, venue quality, volume quality, borrow, and realistic exit liquidity are not. The $159.70 after-hours mark is not a fill guarantee.

I do not have sufficient reliable data to quantify this accurately for a live locate, borrow, or executable exit.

The Mispricing

Fact: Oracle reported $638 billion of RPO, up 363% year over year, and said $75 billion of large AI contracts involved customer prepayment or customer-supplied hardware. Oracle FY26 results

Fact: Oracle reported $31.977 billion of FY26 operating cash flow, $55.663 billion of capital expenditures, and negative $23.686 billion of free cash flow. Oracle FY26 financial tables

Fact: Oracle raised $43 billion of debt financing and $5 billion of equity financing in FY26, and expects approximately $40 billion of debt and equity financing in FY27, including a previously announced $20 billion ATM. Oracle FY26 results

Fact: ORCL closed at $158.78 on September 4, up 3.08% on 26.046 million shares, after closing at $141.32 on September 1. StockAnalysis ORCL history

Inference: The market may be capitalizing the RPO headline and cloud growth before the funding, hardware ownership, capex, margin, and common-denominator consequences are fully reflected in the price.

Reasonable assumption: If Q1 cloud growth is within guide but cash conversion, RPO composition, or funding requirements look worse than the market expects, the stock can give back the September rally toward the $140 base case without an operating collapse.

Unknown: The current split of prepaid versus customer-supplied hardware, contract-level margin and payment terms, Q1 capex and operating cash, actual ATM issuance, the 2026 debt path, customer concentration, and the market's willingness to fund Oracle at the same cost.

The short is therefore a funding-quality disagreement, not an anti-AI claim. Oracle can win the demand argument and still lose the near-term per-share argument if each new dollar of RPO requires more capital, more dilution, or more senior claims than the market's multiple assumes. The market can also be right if customer prepayments and supplied hardware produce superior cash returns while cloud growth stays inside the guide.

The Positioning

MarketBeat reports 47,767,442 shares short as of the August 14 settlement, equal to 2.81% of public float and 1.6 days to cover. Short interest declined 4.74% from the prior report. Those figures are stale relative to the September 10 earnings event and do not identify holders, hedges, or dealer exposure. MarketBeat ORCL short interest

The positioning is not a classic crowded short. There is enough absolute short interest to add selling pressure after a financing or guidance disappointment, but the 1.6 DTC figure means the stock is not structurally trapped. The larger positioning tension is between institutions that can own a high-growth AI proxy and the financing-sensitive holders who must underwrite debt, equity, and customer prepayment risk.

Current positioning score: 3/5. Public short interest is current enough to frame the debate but not current enough to quantify the post-rally state. Borrow fee, utilization, recall and buy-in terms, options dealer gamma, holder concentration, fund flows, and live locate are missing.

The Catalyst

Step	State	Timing	Observable test	Failure mode
Q1 FY27 results	Announced and pending	September 10, 2026 after market close	Check total revenue, cloud growth, IaaS growth, operating margin, EPS, and cash flow	A beat with better margins or stronger guide validates the rally
RPO conversion	Ongoing contract state	Q1 report and subsequent filings	Reconcile new RPO, revenue recognized, contract liabilities, customer prepayments, hardware ownership, and remaining obligations	RPO grows without acceptable margin or cash conversion
Capex and funding	Partly disclosed, not resolved	Q1 report and FY27 financing activity	Check capex, customer-funded offsets, commercial paper, debt issuance, ATM shares, preferred claims, and cash	Funding cost or dilution rises faster than operating cash
Capital return	Discretionary and limited	FY27 filings	Count dividends, repurchases, new shares, preferred conversion, and current denominator	Capital return is overwhelmed by issuance and senior claims
Price acceptance	Unknown	September 11 regular session and following week	Verify opening price, close, volume, spread, depth, venue, borrow, and exit liquidity	A gap or momentum squeeze makes terminal targets unusable

The cheapest falsification sequence is to read the September 10 primary release and tables, reconcile Q1 operating cash and capex against the funding plan, then verify the first regular-session price and borrow state. A clean cloud beat plus lower capital intensity and stable per-share claims invalidates the short. A beat with weaker cash conversion or a higher equity requirement strengthens it, but still does not authorize an entry without the execution audit.

The Payoff

The one stated holding window runs from the first regular session after the September 10 report through the first filing that shows Q1 cash flow and any capital issuance. This is an event-defined window, not a promise to exit on a fixed date.

The top case is $125.00 if cloud growth or RPO quality disappoints, the market removes part of the AI infrastructure premium, and equity funding or margin risk becomes central. The base case is $140.00 if Oracle meets enough of its guide to avoid a collapse but the market gives back the pre-report rally as cash conversion and funding remain unresolved. The bottom case is $190.00 if cloud growth lands above guide, RPO conversion looks high quality, customer prepayments and supplied hardware reduce capital intensity, and momentum buyers absorb the short balance.

Probability rationale: The 30% top case reflects the disclosed FY26 free-cash-flow deficit and the possibility that a large RPO number carries less owner economics than the headline implies. The 45% base case is highest because a mature enterprise company can beat a high-level guide while still failing to resolve the funding-quality question in one quarter. The 25% bottom case reflects the real operating countercase, the September rally, and the market's ability to reward customer-funded AI infrastructure. These are subjective estimates, not observed frequencies.

Price Target and Probability Map

Scenario	Probability	Target / Level	Return / Payoff	Horizon	Conditions	Evidence Quality
Top Case	30%	$125.00	21.27% favorable short move from $158.78	Event-defined through first funding or cash-flow filing	Q1 misses or weakens cloud guide, RPO quality is questioned, capex remains heavy, and equity or senior funding risk is repriced	Medium
Base Case	45%	$140.00	11.83% favorable short move from $158.78	Same	Results are adequate but do not close the capex, cash, ownership, and denominator bridge; the September rally fades	Medium
Bottom Case	25%	$190.00	19.66% adverse rise from $158.78	Same	Cloud beats, RPO converts, customer-funded hardware lowers Oracle's burden, and momentum absorbs the short balance	Medium
Invalidation	n/a	Cloud growth above guide with lower capital intensity and a reconciled per-share funding path	n/a	Same	The funding-quality disagreement is resolved in favor of common equity	High / Medium

Probability-weighted expected value: 0.30 x $125.00 + 0.45 x $140.00 + 0.25 x $190.00 = $148.00, a 6.79% price-only expected decline before costs.

Net expected value: Not responsibly computable. The reference-only map does not contain a verified entry, live borrow, spread, slippage, funding, dividend, gap, or exit-liquidity input.

Current market level and timestamp: $158.78 regular-session close at 2026-09-04 16:00 EDT, with a $159.70 after-hours observation at 19:59 EDT.

Primary instrument: ORCL common stock, unlevered only after every short-execution gate passes.

10/5 favorable base move: 11.83% underlying decline from the regular-session reference.

10/5 credible adverse move: 19.66% underlying rise from the regular-session reference.

10/5 measurement basis: Reference-only, not verified executable entry.

10/5 status: Reject.

Confidence: Medium on the filed cash, capex, RPO, and funding facts; low to medium on target probabilities because the quarter and contract composition are unresolved.

The Kill Shot

The mature counterparty says Oracle has solved the capital problem before it becomes a problem. $638 billion of RPO is a large contractual base, $75 billion is already prepaid or customer-supplied hardware, and Q1 FY27 cloud growth is guided above 58%. Oracle's existing database and applications franchise can provide customer relationships, software attach, and financing access that a pure AI infrastructure start-up would not have. The company generated $32 billion of operating cash flow and can raise capital from a deep market.

That argument defeats a simplistic “negative free cash flow means fraud” thesis. It does not answer whether the current market price compensates common holders for the funding, ownership, margin, and denominator consequences of the buildout. The load-bearing assumption for the short is that the marginal economics of AI capacity will be worse than the market's RPO capitalization implies, or that the cost of funding and dilution will outrun the cash conversion.

The trade can lose even if the long-run concern is right. A Q1 beat can cause a momentum extension before the next funding filing. Customer prepayments can improve working capital. Customer-supplied GPUs can eliminate part of Oracle's capex burden. A high opening price can force a short to cover before the cash bridge is visible. A correct five-year critique is not automatically a profitable one-week short.

What Could Go Wrong

Demand risk: Oracle's cloud and AI demand may continue to exceed its guide. The FY26 release reports cloud infrastructure growth of 77% for the year and 93% in Q4.

RPO quality risk: The $75 billion customer-prepaid or customer-supplied hardware component can reduce the capital Oracle must raise. The undisclosed split and contract economics may favor Oracle more than the bear case assumes.

Cash conversion risk: Operating cash flow can grow rapidly while capex remains high. The market may reward a long-duration buildout before free cash flow turns positive.

Funding access risk: Oracle raised $43 billion of debt and $5 billion of equity in FY26. Its scale and market access may make another funding round less damaging than the short thesis expects.

Denominator risk: The $20 billion ATM is capacity, not issuance. Actual share sales, preferred conversion, and buyback activity can alter per-share math in either direction.

Valuation risk: At a provider forward PE of 19.69 and reported FY26 EPS of $5.83, the stock is not valued like an unprofitable start-up. A moderate multiple can withstand a lot of capex if cloud growth persists. StockAnalysis ORCL statistics

Positioning risk: Only 2.81% of public float is reported short. A small short balance can be covered quickly, and the stock can rise on institutional demand without a classic squeeze.

Gap and halt risk: Earnings can gap beyond any modeled level. A stop order cannot guarantee a 5% exit.

Liquidity risk: The last regular session traded 26.05 million shares, but current bid, ask, spread, depth, venue quality, volume quality, and realistic exit liquidity are not verified.

Instrument risk: No live options chain, open interest, implied volatility, assignment terms, or maximum-loss analysis was performed. Common stock is the only research expression.

What Would Prove This Wrong

The short thesis is invalidated by a reconciled funding-quality improvement:

Q1 total revenue and cloud revenue land at or above guide, with operating margin and EPS intact.
RPO grows through contracts whose revenue, customer prepayments, hardware ownership, and margin terms are disclosed and favorable.
Q1 capex and customer-funded offsets reduce the cash gap without an unexpected equity-denominator increase.
Oracle's next filing shows operating cash flow and free cash flow improving relative to the capacity build, with debt and lease obligations contained.
The first regular-session post-report audit verifies stable price acceptance, current borrow, acceptable spread and depth, and a credible adverse path within 5%.

The fifth condition is an execution gate, not a business conclusion. A strong Oracle quarter can invalidate the thesis and still leave no acceptable short expression.

Risk Audit

Risk	Evidence	Control
Earnings gap	Q1 FY27 results are scheduled for September 10 after close	Keep `entry.price` null until the report is read and the first regular session is audited
RPO and prepayment	$638B RPO and $75B prepaid or customer-supplied hardware	Require the quarterly bridge between RPO, revenue, contract liabilities, hardware ownership, capex, and cash
Funding and dilution	FY26 raised $43B debt and $5B equity; FY27 plans about $40B including a $20B ATM	Track actual issuance, proceeds, preferred claims, debt, commercial paper, capex financing, and current shares
Adverse momentum	$158.78 close after a 12.35% three-session rise	Model $190 bottom; reject if reward to adverse risk remains below 2:1
Borrow and recall	Current locate, fee, utilization, recall, and buy-in terms are missing	Require account-level locate and current lending terms; otherwise do not trade
Liquidity and exit	26.05M shares traded on Sep. 4 but no current depth or exit audit	Verify bid, ask, spread, depth, venue, volume quality, settlement, and realistic cover liquidity
Balance-sheet claims	$31.89B cash and $129.54B borrowings at May 31, with leases and preferred claims	Reconcile cash, debt, leases, preferred, capex, financing, and common denominator before revising targets
Instrument	No options chain or maximum-loss analysis verified	Common stock only; no options, leverage, margin, market orders, or price-floor claims

Best Trade Strategy

Current decision: No executable trade. ORCL is a Reject under the short 10/5 policy and remains a research-only screen.

Primary expression: Unlevered ORCL common stock only after the September 10 result is reviewed and a fresh regular-session quote, broker-level locate, current borrow fee, utilization, recall and buy-in terms, acceptable bid and ask, spread, depth, venue quality, volume quality, settlement, and realistic exit liquidity are verified. The Q1 cash, capex, RPO composition, financing, and denominator bridge must also be available. These are research conditions, not an order.

Reference-only levels: Top $125.00, base $140.00, bottom $190.00. The invalidation state is a favorable RPO-to-cash conversion and funding path, not a mechanical stop.

Horizon: From the first regular session after the Q1 report through the first cash-flow and financing disclosure. The next filing date is not verified.

Do not trade when: the only price is after hours or pre-market; the report is pending; the RPO composition is unknown; actual ATM issuance or customer prepayments are not reconciled; locate, borrow, recall, buy-in, spread, depth, venue, volume quality, settlement, or exit liquidity is unknown; or a Q1 beat creates momentum beyond the $190 stress level.

Alternative: A defined-loss put or put spread could limit gap exposure, but no live option chain, spread, implied volatility, expiry, assignment terms, or maximum-loss analysis was verified. It is not an approved substitute.

Sources

Oracle FY26 earnings release - FY26 revenue, cloud growth, RPO, $75B customer-prepaid or supplied-hardware disclosure, FY26 and FY27 funding, and Q1 FY27 guide, June 10, 2026.
Oracle FY26 investor-relations financial tables - balance sheet, borrowings, leases, cash flow, capex, financing proceeds, preferred stock, share count, and free cash flow.
Oracle FY26 Form 10-K - audited annual filing and common shares outstanding as of June 12, 2026.
Oracle Q1 FY27 earnings date - September 10 after-close release and 4:00 p.m. Central call, September 2, 2026.
Oracle price history - September 1 to 4 regular-session prices, ranges, volume, and after-hours observation.
Oracle statistics and valuation - market capitalization, enterprise value, shares, valuation ratios, and current price context.
Oracle short interest - 47.77M shares short, 2.81% float, 1.6 DTC, and August 14, 2026 settlement history.
Macy's Q1 2026 results - comparison candidate: cash, debt, borrowing capacity, comparable sales, and raised outlook, June 3, 2026.
Macy's Q2 2026 earnings date - comparison candidate: September 10 pre-market report, August 19, 2026.
Macy's price history - comparison candidate: September 4 regular-session close and volume.
Macy's short interest - comparison candidate: 11.66% public float and 6.8 DTC as of August 14, 2026.
Shoe Station Group Q2 earnings date - comparison candidate: September 10 pre-market report and current SHOE listing, August 27, 2026.
Shoe Station Group Q1 2026 results - comparison candidate: sales, margin, cash, debt-free balance sheet, inventory, and guidance, May 21, 2026.
Shoe Station Group price history - comparison candidate: September 4 close, range, volume, and ticker transition.
Shoe Station Group short interest - comparison candidate: 36.80% public float and 13.7 DTC as of August 14, 2026.

Research Quality Scorecard

Criterion	Score	Evidence and deduction
Market disagreement	5/5	$638B RPO and high cloud growth meet a documented negative free-cash-flow and funding bridge
Evidence base	5/5	Fresh earnings-date release, official results and tables, audited 10-K, regular-session market data, and FINRA-derived short data
Positioning and flows	3/5	Public short interest is measurable, but only 2.81% of float; borrow, holder, dealer, and live flow data are missing
Catalyst path	4/5	Q1 report, RPO conversion, capex, funding, denominator, and price acceptance are observable; contract composition is not fully disclosed
Payoff architecture	3/5	Targets and probabilities are explicit, but adverse risk defeats the 10/5 hurdle
Invalidation discipline	4/5	Operating, RPO, funding, denominator, and execution invalidators are monitorable
Differentiated insight	4/5	The key disagreement is contract quality and capital ownership, not an anti-AI slogan
Client value	5/5	The screen shows why a liquid, high-profile setup can still be a bad short before the funding bridge is visible

Total: 33/40. The score supports a substantive no-trade screen. It does not override the 19.66% credible adverse rise, 0.60:1 gross ratio, or missing execution gates.

Bottom Line

Oracle's AI demand is real, its cloud growth is strong, and its $638 billion RPO is not empty marketing. The bear case is narrower: the market may be paying for future capacity before it has seen the return on the capital, the split between customer-funded hardware states, or the common-share consequences of another funding cycle.

ORCL is the best current U.S. short research subject because the September 10 catalyst can test RPO quality, capex, cash conversion, and funding in one report. It is not a qualified trade. The reference price is a regular close, the base price-only decline is only 11.83%, the adverse rise is 19.66%, and the gross ratio is 0.60:1. The Desk's decision is Reject and wait for the primary cash and funding bridge.

AI Illustration Prompt

Create a restrained editorial illustration for The Mispricing Desk: a towering contract ledger marked “RPO $638B” balances on one side of a precision financial scale, while the other side holds a glowing GPU server rack, a red negative free-cash-flow line, and stacked debt and equity certificates. Add a translucent strip labeled “$75B prepaid or customer-supplied hardware” passing underneath the scale, with a small unresolved split mark to show the missing contract composition. In the background, a clean calendar page marked “Sep 10 Q1 FY27” hangs above an understated NYSE price tape moving from $141.32 to $158.78, while a narrow red funding valve suggests the $40B FY27 financing plan. Use graphite, bone paper, deep Oracle-red accents, cobalt blue, brushed aluminum, and sober newsroom lighting; no logos, no cartoon robots, no generic candlestick chart, no dollar-bill imagery, no trading instructions. Leave generous negative space for a headline and include a subtle readable “The Mispricing Desk” watermark in the lower-right corner.