The obvious reading of Intel’s three-day rebound is that Wednesday’s close has restored the bull published on 16 October. It has not. Nasdaq’s official Wednesday close, 28 October 2026, was $95.53, up $1.79, or 1.91 percent, from Tuesday’s $93.74, and only $0.16 above the 15 October close of $95.37. Forty-four sessions through 31 December 2026 put a fresh bull at $122.44, a base at $78.99 and a bear at $58.97. The closes used here are Wednesday’s completed prints. This is not financial advice.
What that older bull did not have to settle is which named shipment would make $122.44 revenue rather than a multiple. On 27 October Intel and Oracle described a plan to expand Xeon computing on Oracle Cloud Infrastructure, including OCI X14 on a processor codenamed Diamond Rapids. The note names no units, no dollars and no delivery date inside 2026. Second-quarter external foundry revenue was $293 million. The third-quarter release is due after Thursday’s close and is not in the formula.
Key facts
- Wednesday 28 October 2026 regular-session close $95.53, up $1.79, or 1.91 percent, from Tuesday’s $93.74. Volume 82,592,440. Open $95.085, high $95.68, low $93.00. The file’s high close is $140.94 on 22 June. Source: Nasdaq historical prices, downloaded 29 October 2026.
- The 20-day average is $100.46 and the 50-day average is $101.65. Wednesday is $4.93 under the first and $6.12 under the second, and $0.55 above the prior five-session high of $94.98. The prior five-session low was $84.645. Source: the same Nasdaq file.
- Forty-four sessions through 31 December 2026: bull $122.44, base $78.99, bear $58.97. The 90-return window starts at the 22 June close. Source: FinanceFeeds calculation from the saved series, 29 October 2026. Not a broker target.
- Intel said on 6 October that third-quarter results come promptly after the close on Thursday 29 October, then a call at 2 p.m. Pacific. Source: Intel’s investor release. The news page, checked 29 October, still listed that call as upcoming. The result was not on it.
- Prepared remarks, 23 July 2026: revenue $16.1 billion. Data Center and AI revenue $6.3 billion, up 59 percent year on year. Foundry revenue $5.8 billion, external revenue $293 million. Third-quarter guide $15.8 billion to $16.8 billion. Capital spending more than $20 billion. Source: Intel’s CEO and CFO remarks.
- On 27 October Intel and Oracle announced plans to expand Xeon computing on Oracle Cloud Infrastructure, including planned OCI X14 on a processor codenamed Diamond Rapids. The release states no units and no dollars. Source: Intel’s release. The same plan, still with no units, is in Intel’s post.
- Karan Batta wrote on 27 October that planned OCI X14 offerings use Diamond Rapids, that Google TPUs are planned on OCI from 2027, and that OCI cloud infrastructure revenue was $7.4 billion in the first quarter of fiscal 2027, up 121 percent. Source: Oracle.
- Intel said on 26 October that it will acquire Classiq, with a close expected in the first half of 2027. The release states no price. Source: Intel’s Classiq release. Investing.com reported that Harlan Sur at JPMorgan moved Intel to Neutral from Underweight and kept an $85 target. Source: Yahoo Finance.
What just happened, and why the obvious reading is wrong
Wednesday was the third up close in a row, not a return to June. From Friday’s $86.96 to Wednesday’s $95.53 is $8.57, or 9.86 percent. Thursday 22 October closed at $85.49, down $7.47, or 8.04 percent, on 134,128,100 shares, with a low of $84.645.
A reader who sets $95.53 next to $95.37 will think the old ten-session bull of $117.12 and bear of $80.96 still apply. The price is $0.16 higher. The distribution is not. Those figures used ten sessions and a less negative drift. This path uses 44 sessions, and the 90 returns now begin at the 22 June close of $140.94. The mean of that window is negative, so the drift-only base is $78.99.
The close is not a breakout just because it cleared a five-session high. Wednesday finished $0.55 above the prior five-session intraday high of $94.98, and $4.93, or 4.91 percent, under the 20-day mean and $6.12, or 6.02 percent, under the 50-day mean. It is down $45.41, or 32.2 percent, from the 22 June close. Only that decline is inside the formula.
FinanceFeeds recorded the 21 September jump from $108.60 to $121.78. That session is the largest absolute log-return in the 90-day window, 0.114545, and 8.18 percent of the sum of squared deviations. It does not dominate sigma. The negative mean is the path down from $140.94.
Who is exposed
Oracle is the named CPU counterparty this week, and the paper is a plan. Intel’s 27 October release says the companies intend to coordinate roadmaps and capacity planning, and that Diamond Rapids-based OCI X14 offerings are expected. Batta’s note also lists Nvidia, upcoming AMD instances, and Google TPUs from 2027. Neither note books a unit.
The nearer exposure is the Xeon Intel already sells. Dave Zinsner said on 23 July that server CPU demand continues to far outpace supply, with supply growth skewed toward the end of the third quarter and into the fourth. Data Center and AI revenue was $6.3 billion, up 24 percent sequentially, with operating profit of $2.5 billion. A shipment that can hit the books by 31 December is a current Xeon, not the next-generation part in the 27 October note.
Foundry is the smaller line. Of $5.8 billion in foundry revenue, $293 million was external, 5.1 percent of foundry and 1.8 percent of the $16.1 billion company total. Tan named a Fortinet security-processor collaboration as an ASIC step and attached no dollar. On 14A he cited customer engagements and named no volume customer. An earlier FinanceFeeds note asked that customer question before Intel set 29 October.
Wednesday’s $95.53 is $10.53 above the $85 JPMorgan kept, and that $85 is $6.01 above the $78.99 base. Investing.com reported server x86 share down from 77 percent to 53 percent, and PC share from 81 percent to 68 percent, since the bank’s 2022 downgrade. Those figures are the article’s, not a new Intel filing. Classiq does not close until the first half of 2027.
What the tape and the documents show
Change is versus the prior close in the Nasdaq file. Thursday 29 October had no cash bar in that file. The Nasdaq info endpoint, saved the same morning while the session was still premarket, is not the spot.
| Session | Close | Change | Volume |
|---|---|---|---|
| Wed 28 Oct 2026 | $95.53 | +$1.79, +1.91% | 82,592,440 |
| Tue 27 Oct 2026 | $93.74 | +$3.27, +3.61% | 90,523,290 |
| Mon 26 Oct 2026 | $90.47 | +$3.51, +4.04% | 98,744,160 |
| Thu 22 Oct 2026 | $85.49 | -$7.47, -8.04% | 134,128,100 |
| Thu 15 Oct 2026 | $95.37 | prior house spot | 147,157,700 |
| Mon 21 Sep 2026 | $121.78 | +$13.18, +12.14% | 191,638,400 |
| Mon 22 Jun 2026 | $140.94 | high close in the file | 127,190,600 |
| Wed 31 Dec 2025 | $36.90 | year-end close | 51,524,680 |
The 16 October close is $89.85, 36.2 percent under $140.94, the Friday behind Investing.com’s about 36 percent line. Wednesday is 32.2 percent under that June close. The info file’s 52-week high of $142.35 is the 30 June intraday print, not a close.
| Number | Price | Versus $95.53 | What it is |
|---|---|---|---|
| Bull used here | $122.44 | +28.2% | 44 sessions, drift plus 1.5 standard deviations |
| Base | $78.99 | -17.3% | Same 44 sessions, drift only |
| Bear used here | $58.97 | -38.3% | Drift minus one standard deviation |
| JPMorgan, Harlan Sur | $85 | -11.0% | Neutral, target unchanged, as reported by Investing.com |
| Prior house pair | $117.12 / $80.96 | off $95.37 | Ten sessions on 16 October. Not this path |
| Ten-session check | $112.75 / $79.59 | +18.0% / -16.7% | Same formula, N of 10, through 11 November. Not used |
FinanceFeeds has already set out the second-quarter loss under the July beat. That loss is not an input here. The bull of $122.44 is $18.50 under the June high close and $37.44 above Sur’s $85.
What Intel, Oracle and the banks have said
“Today, we are seeing the strongest revenue growth in more than fifteen years,” Lip-Bu Tan, chief executive, said in the 23 July remarks. He also said, “We also began risk production of 18A-P.” On 14A he said PDK 0.9 was on track for October, internal risk production was the second half of 2027, and high-volume ramps were committed for 2028. He named no external 14A customer.
“Due to strong customer demand signals, we are raising our outlook for 2026 and now expect our cap ex to be more than $20 billion, which is up significantly versus our expectations entering the year,” Dave Zinsner, chief financial officer, said in the same remarks. He said 2027 spending would be significantly higher, mostly in the United States, and that US tools and space from 2021 through 2026 were approaching $100 billion. External foundry revenue in his section was $293 million.
“Our expanded collaboration with Oracle is intended to combine Intel’s multigenerational roadmap with Oracle Cloud Infrastructure’s technology to help customers put AI to work faster and more cost-effectively,” said Kevork Kechichian, executive vice president and general manager, Intel Data Center Group, in Intel’s 27 October release. “Enterprise AI depends on the applications, data, and tools that let an agent complete useful work,” said Pradeep Vincent, executive vice president, compute and AI infrastructure, Oracle Cloud Infrastructure, in the same release.
“I am delighted to have Nir, Yehuda, Amir and the entire Classiq team join Intel,” Tan said in the 26 October release. His post the same day says: “With Classiq, Intel will become a full-stack quantum solutions provider and add QPUs to our already diverse compute platform portfolio spanning CPUs, GPUs and ASICs.” A 28 October post carries him, as that account presents it, quoting an unnamed customer: “And they told me, ‘Lip-Bu, you guys screwed up 15 areas.'” The post names no firm and no node.
“Risk/reward is more balanced at current levels and we move to a Neutral rating,” Harlan Sur said, as reported by Investing.com. Investing.com also reported that he expects agentic-AI workloads and tight leading-edge supply to support earnings revisions through 2027, and that the bank’s 2028 estimate is about 15 percent above a consensus figure the page does not print. FinanceFeeds covered that upgrade as news. His $85 is separate from the $58.97 bear.
The spending is real, and the invoice is not
The external foundry invoice in the same script is $293 million. Investing.com reported barely $1 billion of annualized external foundry revenue in the first half, against $150 billion at TSMC. Four times $293 million is $1.172 billion, the run-rate behind that line. Cash and short-term investments were approximately $30 billion, which is liquidity, not a Classiq price. The release states no price. The 14A PDK 0.9, on track for October in Tan’s July remarks, confirms a process date. It does not book an Oracle unit.
Bull, base and bear
Sur’s $85 is his target. The title figures are the distribution.
The last 90 daily log-returns run from the 22 June close of $140.94 through Wednesday. The mean, mu, is -0.004321043707. The sample standard deviation, sigma, dividing by 89, is 0.044052998929. Forty-four sessions remain through 31 December: 29 and 30 October, 20 November weekdays after dropping Thanksgiving on 26 November, and 22 December weekdays after dropping Christmas on 25 December. The early closes on 27 November and 24 December count. 2 plus 20 plus 22 is 44.
The square root of 44 is 6.633249581. Sigma times that root is 0.292214537, and 1.5 times the band is 0.438321805. Forty-four times mu is -0.190125923. Base is $95.53 times e to the power of -0.190125923, or $78.98946, which rounds half up to $78.99, down 17.3 percent. Bull is $95.53 times e to the power of 0.248195882, or $122.44185, which rounds to $122.44, up 28.2 percent. Bear is $95.53 times e to the power of -0.482340460, or $58.97419, which rounds to $58.97, down 38.3 percent. The negative drift does not pull the bull under the spot, so the bull keeps the drift.
Ten sessions through 11 November round to $112.75, $91.49 and $79.59. Fifteen sessions through 18 November round to $115.65, $89.53 and $75.49. The year-end bull is $5.32 from $117.12 and $22.56 from $145. The year-end bear is $3.03 from $62 and $3.97 from $55. The collision rule is $1, so this pair is the title. A 20-session bull of $117.75 does not clear that rule and is not used.
| Case | Price | Versus $95.53 | What has to be true |
|---|---|---|---|
| Bull | $122.44 | +28.2% | Drift continues and 31 December finishes 1.5 standard deviations above it. It is revenue only if a named Oracle Xeon shipment or a named external foundry customer is in the books. |
| Base | $78.99 | -17.3% | Only the mean continues for 44 sessions. No new customer is required. The path is the decline from $140.94, not the gain from $36.90. |
| Bear | $58.97 | -38.3% | The drift holds and the close finishes one standard deviation under it. A guide miss can travel this way. Sur’s $85 is separate, and $26.03 above this bear. |
$122.44 is a multiple unless the books name the shipment. Diamond Rapids on OCI X14 is not that shipment until Oracle or Intel puts units, dollars or a 2026 date on it. Current Xeon, the processor Zinsner says is still short, is what could show up before 31 December.
What happens next
On Thursday 29 October, after the cash close, Intel reports the quarter guided to $15.8 billion to $16.8 billion. The call is at 2 p.m. Pacific, so that cash close cannot contain the print. If the release names no Oracle unit and no external 14A customer, $122.44 stays a distribution. A print inside the guide does not, by itself, move the bull.
Friday 30 October is the first regular session that can price the call. A revenue print under $15.8 billion, or a cut to the more-than-$20-billion spending line without a named shipment, is the sort of fact that can travel toward the $58.97 bear. A beat of the $16.3 billion midpoint still has to do the work of 1.5 standard deviations over the sessions that remain. One session cannot finish a 44-session path.
By 31 December the test is whether Data Center and AI names an Oracle Xeon shipment, or external foundry names a customer above the $293 million line. Classiq cannot close before the first half of 2027, and Batta’s Google TPUs start in 2027. If 31 October passes with no word on the 14A PDK 0.9, the July sentence has slipped. Until one of those revenue sentences exists, $122.44 is the upper band, not an invoice.
Frequently asked questions
Where do the sessions through 31 December 2026 put Intel, and what has to show up for the bull to be revenue?
Forty-four sessions from Wednesday’s $95.53 Nasdaq close put the bull at $122.44 and the bear at $58.97, with a base at $78.99. That bull is a revenue outcome only if a named Oracle Xeon shipment or a named external foundry customer shows up in the books. The 27 October Oracle note names Diamond Rapids and OCI X14, and it names no units and no dollars. Second-quarter external foundry revenue was $293 million.
What bull and bear does this Intel stock prediction use?
The bull is $122.44 and the bear is $58.97, with a base of $78.99, off the $95.53 close. They use 44 sessions through 31 December 2026, the mean and the sample standard deviation of the last 90 daily log-returns, and bands of 1.5 and 1.0 standard deviations. The mean is negative because the window starts at the 22 June close of $140.94. These figures are not JPMorgan’s $85 target.
When does Intel report, and can Thursday’s close contain the print?
Intel said on 6 October that it will report third-quarter results promptly after the close on Thursday 29 October, then hold a call at 2 p.m. Pacific. The cash close that day cannot contain the print. The July guide is $15.8 billion to $16.8 billion, with non-GAAP earnings of $0.38 at the $16.3 billion midpoint. Friday 30 October is the first regular session that can trade the release.
What did Oracle and Intel actually commit?
They committed to a plan, not to a counted shipment. Intel’s 27 October release says the companies intend to coordinate roadmaps and processor capacity planning, and that Diamond Rapids-based OCI X14 offerings are expected. Pradeep Vincent said the collaboration is intended to grow compute capacity. Intel’s post the same week uses the words plans and planned. No unit figure is in either text.
Does the Classiq deal change the 2026 revenue path?
No. Intel said on 26 October that the acquisition is expected to close in the first half of 2027, subject to regulatory approvals, and that the companies stay separate until then. The release states no price. Lip-Bu Tan’s post the same day calls the deal a way to add quantum processors to a portfolio of CPUs, GPUs and ASICs. That is not a 2026 Xeon shipment and it is not external foundry revenue.
Why is the base under the close if the stock is up on the year?
The base uses the last 90 daily log-returns, not the move from 31 December 2025. Those returns begin at the 22 June close of $140.94, the high close in the file, and the mean is negative. Wednesday’s $95.53 is still up $58.63, or 158.9 percent, from the $36.90 close on 31 December 2025. The year-end base of $78.99 is what that recent decline does if it continues for 44 sessions.
