Is Oracle undervalued at $140? My DCF says yes, but the balance sheet is the real risk

A few months back I posted [Oracle](https://www.reddit.com/r/ValueInvesting/comments/1s9513t/oracles_60_crash_cheap_value_play_or_leveraged_ai/) analysis here and was cautious. It is around $141 now, roughly at the same price, but it got there the hard way. It ran up to roughly $200 after a record June quarter, then gave all of it back. A lot of the softer signals have improved, but the one risk I actually flagged has gotten worse rather than better. So I thought it was time for a quick update.

First the good news. I wrote:

"But the bull case is built entirely on analyst estimates. The revenue doubling requires a 40% CAGR over the forecast period. Oracle's actual 9-year historical revenue CAGR is 5%. You're being asked to believe the growth rate will be eight times its historical average, sustained for years, based on a backlog that hasn't yet converted to recognised revenue."

\- So far the analysts have been right. FY 2026 revenue came in almost exactly on their estimate. So I was being over cautious.

"On top of this, insiders, including the CEO, CFO, and multiple presidents, have sold more than $3 billion in shares over the past eight quarters. Purchases over the same period: roughly $1 million. The people with the clearest view of whether that backlog actually converts are not buying."

\- This selling has now stopped or reversed.

"The bull case is real, on paper. Oracle holds $553 billion in remaining performance obligations - contracted future revenue"

\- The backlog has since grown to $638 billion. And OpenAI's $122B raise takes some pressure off the funding question behind roughly half of it.

Now the deterioration. My original post says:

"The balance sheet makes this harder to ignore. Oracle is carrying $104 billion in debt, a debt-to-equity ratio above 5x, and free cash flow that turned negative last year after averaging $12 billion annually for nearly a decade. Across 80 large-cap tech peers, Oracle sits in the bottom 5th percentile on leverage. No other major software company has stretched its balance sheet this far while simultaneously betting on a growth inflection."

\- This is where it's gone the wrong way. On 8 July, S&P cut Oracle to BBB- and the numbers behind it aren't pretty. Total debt sits around $167 billion and is still climbing. Oracle is raising capital hard to fund the buildout: a $5 billion convertible preferred in February, a $20 billion equity issuance planned this year, and tens of billions more flagged over the next three years.

"A standard DCF spits out $300 intrinsic value, implying massive upside."
"Analysts project revenue nearly doubling from $57 billion today to $130 billion by FY2028."

\- The valuation math is broadly unchanged. It still runs on analyst revenue estimates, now around 30% a year, with revenue tripling to roughly $218 billion by FY2030. The difference is that the first year of that ramp actually happened, so those estimates carry more credibility than they did when I first wrote about it.

So where does that leave us? The demand is real and the backlog is enormous, but the whole thesis now rests on execution. Our DCF still says the reward is there at roughly $300 against $141 today. But that number assumes clean execution, and the S&P cut is the market reminding you that clean execution is exactly what's in doubt. Cheap for a reason, or cheap despite a fixable problem. That's the call you're actually making here.

Disclaimer: This is for educational purposes only and is not investment advice. The author and Stockoscope may hold positions in the securities mentioned. Always do your own research.