At The Motley Fool, we poke plenty of fun at Wall Street analysts and their endless cycle of upgrades, downgrades, and "initiating coverage at neutral." So you might think we'd be the last people to give virtual ink to such "news." And we would be -- if that were all we were doing.

But in "This Just In," we don't simply tell you what the analysts said. We'll also show you whether they know what they're talking about. To help, we've enlisted Motley Fool CAPS, our tool for rating stocks and analysts alike. With CAPS, we track the long-term performance of Wall Street's best and brightest -- and its worst and sorriest, too.

Sounds of silence
Speaking of the worst, let's pause today and enjoy a blessed moment of silence as Goldman Sachs quietly mulls the mistake it made in urging investors to buy Nuance Communications (Nasdaq: NUAN) last month. [Cue crickets, chirruping.] 

As you've probably heard by now, Nuance reported third-quarter earnings Monday evening, and the news wasn't pretty. The speech-to-text technologist missed consensus estimates on sales, and guided low, or at least not high, for this quarter and full year. The low end on this quarter's guidance suggests as much as a $15 million shortfall on Wall Street's hoped-for revenue number, alongside a potential earnings miss of $0.02. 

Investors, having sold the stock off by more than 11% already, do not appear pleased -- and neither is Goldman. Doing an abrupt about-face, Goldman quickly pulled its "buy" rating and downgraded the stock to "neutral." And yet, after reviewing the rating, I'm hard-pressed to see why. 

After all, while it's dialed back expectations in response to Nuance's less-than-ebullient projections, Goldman still says it expects to see $1.19 per share in profit at Nuance this year, $1.27 in 2011, and $1.38 in 2012. Combined, the analyst's revisions add up to roughly just a 4% reduction in predicted profits over the next couple of years. 

Four percent and a cloud of dust
Four percent? Is that really the margin of error Goldman incorporates into its "buy" and "neutral" ratings? Sheesh! Personally, I don't buy anything unless -- by my admittedly rough valuation method -- I see at least a 20% margin of safety. Yet Goldman appears to think it's got valuation down to such a science that it can say "X equals hold, X plus 4% equals buy," and hop in and out of stocks nimbly as they cross and recross that threshold. 

Which would be great for Goldman, except for one thing: This banker doesn't seem to be following its own rules. One month ago, Goldman laid out its buy thesis for Nuance thusly: Nuance was selling for 11 times fiscal 2011 earnings. In contrast, across the length and breadth of the software sector, similar stocks commanded an average 16 multiple to such forward earnings. 

Never mind that I examined the analyst's argument last month and found it wanting. While it was true that Microsoft (Nasdaq: MSFT), Adobe (Nasdaq: ADBE), and Oracle (Nasdaq: ORCL) all sold for close to Nuance's then-11-times ratio, and still do (more or less). The only way to argue that "16 times" was the norm in this sector required averaging in such valuation outliers as Akamai (Nasdaq: AKAM), VMware (NYSE: VMW), and salesforce.com (NYSE: CRM), which today range in price from 27 times forward earnings (Akamai) all the way up to 67 times (salesforce).

In other words, Goldman's 16-times hypothesis on Nuance simply wasn't credible, which is why I panned the pick

Eat your cake or have your cake. Pick one.
Let's re-examine Nuance today under the harsh light of Goldman's original hypothesis. Assume, first of all, that Nuance would be "a buy" at 11 times 2011 earnings. If Goldman's right about $1.27 being the correct number for Nuance's 2011 profits, this would imply the stock will become cheap enough to merit a buy rating at about $14 a share. 

Very well. At $15.50 today, it's not quite there yet. But what about the stock's fair value? What would Goldman say about that today? Assuming the analyst is sticking with its 24-karat guns, you'd expect Goldman to argue in favor of a $19.50-a-share price on Nuance today, which is where it pegged it last month, right? Wrong. Care to guess what Goldman actually thinks the shares are worth, today? 

Answer: $17.50. 

Foolish takeaway
Crazy as it may sound, Goldman has taken a 4% projected shortfall in Nuance's profit and turned it into about a 10% reduction in the stock's estimated value. In Goldman's mind, this probably all seems quite logical, maybe even scientific.

To me, though, it just confirms what I wrote last month: The analyst didn't demand a big enough margin of safety on Nuance. And now, investors who listened to Goldman are paying the price.