How did the community of analysts–who usually are pretty good about this sort of thing–miss their forecast of Intel’s sales by more than a billion dollars? How did they also miss Intel’s earnings per share by 13 cents? Is it really a case of what Intel CFO Stacy Smith referred to on a conference call last night as two blind people describing different parts of an elephant? Analyst Doug Freedman with Gleacher and Co. tried to explain it in a note to his clients today. He says two things contributed to the mass misjudgement.

First, everyone misjudged the number of chips consumed in the sales channel during the third and fourth quarters of 2010 ahead of Intel’s big launch of its latest Sandy Bridge generation of PC chips in January of this year. Having burned through their inventory of older chips, Intel shipped more of the newer chips than most people had expected, as distributors bulked up their stocks after cutting them down more than usual. Freedman says he thinks the same thing is still going on, which means that the guidance that Intel gave for the second quarter may turn out to be conservative.

Second, there was an extra week during the quarter. If you listened to the conference call, you heard several references to Q1 having 14 weeks. If you add a week’s worth of revenue–about $818 million–plus the additional revenue brought in by Intel’s acquisitions of McAfee and the Infineon wireless unit, which works out to another combined $496 million, you end up with about $1.3 billion, which is almost exactly the difference between the Wall Street consensus revenue number and Intel’s actual revenue.

Whatever the reason for the disconnect between the forecasts and the results, Freedman boosted his price target on Intel to $28 from $27 and maintained his Buy rating. Even so, other analysts weren’t quite so willing to change their stripes. Michael McConnel of Pacific Crest Securities, continued to insist that the slowing market seen by Gartner–who Intel’s Paul Otellini criticized in not-so-veiled comments during the conference call last night–will catch up to Intel sooner or later. Another, Stacy Rasgon of Sanford Bernstein, wrote that he’s not buying Intel’s story, period. Tiernan Ray summarized the day’s scorecard on revisions and recriminations by Intel analysts.

Whatever they said, Intel’s share price roared by nearly 8 percent to close at $21.41, the highest levels it has seen since March. Never mind all that pessimism from yesterday. Come on, analysts, it’s time to work your Emily Litella impersonations. Say it with me now: Never Mind!

Here’s a clip from Saturday Night Live circa 1975, courtesy of Hulu, to get you started. Better luck next quarter.

“Our focus is to build our [intellectual property] portfolio,” he said during a meeting with analysts Thursday. “You could see us buying chip companies. Silicon is very important.”

So if Oracle (ORCL) were to acquire a chip company, which would it be? Wall Street seems to think it’s ARM. The chip designer’s shares spiked about six percent following Ellison’s remarks, though it’s not clear that ARM is what he had in mind when he made them. Certainly there are other potential targets: Broadcom, Marvell (MRVL), Nvidia (NVDA) and AMD (AMD), of course. Said Gleacher & Co. analyst Doug Freedman, “You’ve got to think it’s focused on enterprise hardware, on the server. AMD jumps off the screen.”

He notes that of the 19 companies he follows, 14 have posted gains of 5% to 25% since he upgraded the sector on April 4. (Just two of those stocks–Qimonda (QI) and Omnivision (OVTI)–have declined in the subsequent weeks.) And now, he says, they are due for a four- to six-week pause.

Read the rest of this post