D.A. Davidson’s Gil Luria sees transformational upside in Oracle (NYSE: ORCL), telling investors on a Nov. 16 podcast appearance that “if they execute perfectly, the stock is a multiple of where it is today.”
At 14 times forward earnings, Oracle is trading well beneath its five-year average multiple of 21 times, a gap that reflects deep skepticism about whether its massive contract backlog will ever materialize into free cash flow.
The arithmetic behind Luria’s thesis is straightforward: if Oracle reaches its own target of $21 or more in non-GAAP earnings per share by fiscal 2030 and the multiple stays at today’s compressed level, the stock would be worth roughly $294 per share.
That figure represents more than 2.5 times Oracle’s Nov. 17 closing price of $115, and even the Wall Street consensus target of approximately $236 implies the stock could roughly double from current levels.
The problem is that Oracle’s path to those numbers runs directly through a free cash flow deficit that reached negative $23.7 billion in fiscal 2026, with Street analysts projecting only $17.5 billion for fiscal 2029 against Oracle’s own $30 billion target.
Oracle set out its financial roadmap at an Investor Day on Oct. 28, capping new capital raises at $20 billion through fiscal 2030, then proceeded to deploy that entire ceiling within five days by pricing 86.2 million new shares at $116 each and raising $10 billion in notes on Nov. 2.
That move clears the financing overhang but leaves Oracle with little room for cost overruns, which must now be absorbed by customers, suppliers, or Oracle’s own balance sheet.
Two contract setbacks followed in quick succession: the Financial Times reported on Nov. 4 that Oracle had walked away from a roughly $7 billion chip-leasing arrangement with Tencent after U.S. officials raised concerns, and Bloomberg reported two days later that Microsoft is allowing a 2023 agreement to rent older Nvidia chips from Oracle to expire.
Neither loss punches a significant hole in Oracle’s finances, since the Tencent deal amounted to roughly 1% of the $664 billion backlog, and the outgoing GPU contracts had been priced at a 20% premium to market rates just last quarter.
The far larger risk is concentration in OpenAI, with commentator Steve Eisman estimating that approximately half of Oracle’s entire backlog is tied to a single customer whose own financing picture grew more complicated after Nvidia’s then-finance chief Colette Kress reportedly pushed back on a proposed $250 billion credit backstop for the artificial intelligence company.
CoreWeave offered a partial counterbalance, beating third-quarter expectations on Nov. 10 and raising its full-year guidance, sending Oracle shares up 8.8% the following session, though CoreWeave also disclosed that new contracts are pricing at $20 million to $25 million per megawatt annually, compared with the roughly $16 million figure Oracle has cited.
Oracle’s broader selloff, a 10% decline in the month through Nov. 17, tracked a wider retreat in AI infrastructure names, with Nvidia falling 27% and CoreWeave dropping 15% over the same period while the Nasdaq 100 shed 7.5%.
The December quarterly report now serves as the immediate checkpoint, with Oracle having guided for revenue growth of 30% to 34% and non-GAAP earnings per share of $1.85 to $1.93, and investors watching closely to confirm that the Tencent and Microsoft exits left no visible gap in the backlog.