FORTUNE — Consider the research analyst. Toiling in a job on Wall Street that is at once one of its most visible and least glamorous. Expected to write with integrity while serving – all too often – at the pleasure of the underwriting and trading desks. Forced into a shotgun marriage between the cold logic of fundamental analysis and the surreal dreamlife of a speculative bubble.

Many analysts, of course, may avoid this fate if they follow bread-and-butter sectors like chemicals or machine equipment. Life is trickier for those covering Internet stocks, especially in markets where there’s money to be made from an imminent bubble in web IPOs. Then, the research analyst must resort to a logic more familiar with medieval alchemists, whose task was to extract pure gold from something that everyone else knows is just an ordinary rock.

We are starting to see this alchemy 2.0 now that LinkedIn’s underwriters are issuing reports on the stock. LinkedIn (LNKD) went public on May 19 at $45 a share. It closed its first day at $94 and rose as high as $107 the next day. Then just as suddenly, the hot air started to leak out of the balloon and LinkedIn shares fell as low as $60 last week.

On Tuesday, cheery analysts employed by LinkedIn’s underwriters showed up with fresh propane for LinkedIn’s hot-air balloon. Most of their reports had a kind of bipolar feel to them. On the one hand, they rightly pointed out that LinkedIn is a well-run company positioned to draw profits from its social networks for professionals. On the other, they tried to justify its post-IPO stock price, which is 27 times its sales and 510 times it earnings over the past 12 months.

To be clear, the question isn’t whether LinkedIn will fail. It’s whether LinkedIn is worth buying right now. And the analysts at LinkedIn’s lead underwriters have answered that question with a full-throated Yes!

JP Morgan gave the stock an “overweight” rating – Wall Street’s version of a meddling aunt who thinks you should really eat more – and put an $85 price target on it. Morgan Stanley slapped a “buy” rating on the stock and envisioned a price of $88 in its future. BofA Merrill (BAC) saw that price target and raised it to $92.

And somehow, it worked. On Friday, before these reports were published, LinkedIn’s stock was trading at $70 a share. By Tuesday’s close, the reports helped lift the stock back above $85.

To accomplish that, the underwriters’ analysts offered a spicy word salad: “transformative,” “viral,” “disruptive” and “penetration.” Excited yet? How about this: JP Morgan wrote, “Much like Facebook is doing in social, LinkedIn is mapping the professional graph.” And Morgan Stanley crowed that LinkedIn is “empowering professionals in a connected world” – which sound like some rather disruptive penetration, until you remember that that’s what email has been doing for pretty much the last 20 years.

For all that, LinkedIn is expected to lose between $19 million and $25 million this year on a GAAP basis. But these analysts see the company returning to a profit in 2012. And in the meantime, they are valuing the company not on price-to-sales or price-to-earnings but on something called DCF. To you and me, that means discounted cash flows.

DCF is a metric that might strike individual investors as arcane. It’s actually somewhat common – it compares the current value of a property against the cash flows it’s expected to produce in the future. But the thing is, it’s usually reserved for conservative investments. It isn’t invoked in speculative sectors like the web unless people are kind of desperate.

And LinkedIn’s future cash flows are fairly speculative. But let’s say they’re not. Let’s forget for a moment that LinkedIn is spending its IPO proceeds to leverage future growth at the expense of near-term profits. Or that, to achieve that growth, it’s is starting to milk its users for revenue in ways that they may not like. Or that there are several stormy clouds on the global horizon that could put a dent in LinkedIn’s growth curve.

Even so, LinkedIn’s underwriters are setting its DCF high. Morgan Stanley (MS) argued that LinkedIn’s DCF valued the stock at $88 a share. JP Morgan (JPM) thought it meant the stock could go as high as $100. But on June 13, when the stock was trading at $72 a share, an independent firm, Evercore Partners (EVR), argued that LinkedIn’s DCF valued the stock as low as $59 a share. Evercore’s price target at the time was $70. And on Tuesday, another independent firm, Montrose Securities, asserted LinkedIn’s fair value was $62 a share.

The other question that these bullish analyst reports raise is this: If they really think LinkedIn is worth $85 or more a share, why did their own firms price the IPO at $45? That only adds credence to skeptics like Joe Nocera, who said LinkedIn was leaving hundreds of millions of dollars on the table with that low offering price.

Unless, of course, LinkedIn wanted to leave that money on the table. And why would it do that? I can’t think of a good reason, except that LinkedIn believed listing a mere 10% of its shares would create an imbalance between demand and supply – which would allow it to offer a larger portion of shares later on at an artificially inflated price. (Bizarrely, JP Morgan listed the small float as a positive for LinkedIn’s stock price.)

Of course, LinkedIn – and its underwriters – would never consider that option if they’re being honest. After all, an ordinary rock isn’t gold. It will always be just a rock. It’s strange to think that people can sometimes confuse the two. But isn’t that why we have analysts? It’s their job to separate the real gold from the fake.