ORCL posted its strongest single-day gain in weeks on Tuesday, up 4.7%, yet the week as a whole remains slightly negative — and the options market, which was screaming bullish two weeks ago, has quietly retreated toward neutral.
The options angle deserves close attention, because two consecutive notes flagged the call-heavy positioning as the standout divergence in the dataset. That signal has faded. The put/call ratio is now 0.84, back inside the 20-day average of 0.92 — just one standard deviation below it. Three weeks ago the PCR hit 0.76, more than two standard deviations below average, the most call-heavy reading in nearly a year. That extreme has unwound. The message from options traders is no longer "we expect a sharp recovery." It is closer to "we are watching."
The borrow market offers nothing of interest to shorts. Availability is essentially unlimited — the lending pool dwarfs the shares borrowed by a factor of roughly 100-to-1, and cost to borrow has dropped nearly 46% over the past week to just 0.22%, the lowest level in the 30-day history. Short interest remains light at 1.35% of the free float, marginally lower on the week. There is no evidence of a crowded short position, no squeeze pressure, and no cost friction that would discourage new short sellers from entering. The borrow market is a non-story.
Where the Street sits is the more interesting tension. CLSA initiated coverage on Monday with a Hold and a $145 target — almost exactly where the stock is trading now, and well below the analyst consensus mean of roughly $248. That gap is extraordinary. The bulls at Guggenheim ($400 target), TD Cowen ($300), and Bernstein ($325) are pricing in the AI infrastructure thesis: 41% cloud revenue growth, 243% growth in AI workloads, a $90 billion FY27 revenue guide, and remaining performance obligations that exploded 325% year-over-year. The bears, led by CLSA's fresh-eyes initiation, are pricing in a company already at fair value after a 31% one-month decline. The EP (earnings yield) factor score ranks in the 94th percentile for analyst recommendation divergence — the Street has rarely been this split on a name this size. Valuation multiples have compressed sharply: the trailing PE has dropped 5.4 turns over 30 days to 13.2x, and price-to-book has shed 1.6 turns to 3.7x. On those measures, Oracle is cheap relative to its own history.
Peer performance this week adds one more wrinkle. CDNS fell 8.5% on the week — actually worse than ORCL's 0.7% decline. RBRK dropped 13.9%. DDOG lost nearly 6%. MSFT, by contrast, gained 3.3%. The picture is no longer "Oracle is being re-rated in isolation while peers rally." Selling pressure has broadened across enterprise software this week, which complicates the ORCL-specific narrative that dominated the prior six weeks. The exception is Microsoft, which continues to track its own AI-driven re-rating upward — and which represents precisely the competitive pressure Oracle's bears cite.
The next scheduled earnings event is September 4. Between now and then, the question is whether Tuesday's bounce represents the beginning of stabilisation at current levels or a reflexive bid in an ongoing downtrend — and the cooling of the options bull signal means that question is, for now, genuinely open.
See the live data behind this article on ORTEX.
Open ORCL on ORTEX →ORTEX Market Intelligence content is generated by AI from a snapshot of ORTEX's proprietary data. Content is informational only and does not constitute investment advice.
