This series, brought to you by Yahoo! Finance, looks at which upgrades and downgrades make sense and which ones investors should act on. This morning, with markets deep in the red, we're looking at a series of downbeat moves on the Street as analysts "upgrade" EnerNOC (Nasdaq: ENOC) and knock a few zeroes off the market caps of Apache (NYSE: APA) and General Motors (NYSE: GM). Let's dig in.

You call this an upgrade?
One thing you need to know if you're going to invest in the stock market is that the advice Wall Street gives you on the stocks it rates doesn't always…make sense. Any sense at all. Take, for example, the "upgrade" that Ardour Capital just assigned to smart-grid specialist EnerNOC.

It's clear Ardour has feelings for EnerNOC, as it now endorses the shares. And yet these feelings are just as clearly conflicted. At the same time it recommended "accumulating" the shares (presumably by rummaging through the bargain bins at your neighborhood garage sale), Ardour cut 25% off its price target on the stock, which it now values at $7.50 per share.

Why the mixed signals? Well, on the one hand, EnerNoc reported first-quarter results this morning, showing "margins trending higher" and "guidance for 2012 increased." That sure sounds good. On the other hand, the company still managed to lose $1.06 per share. EnerNOC also admits it's likely to lose money over the course of the full year -- as much as $1.50 per share in total.

So is there any reason to follow Ardour's advice and buy this stock? Perhaps as a turnaround play. Losses aside, EnerNOC does have more than $80 million cash in the bank (more than half its market cap) and zero debt. Cash burn is minimal, so if you think the business is worth at least $3 a share, the cash alone makes up the difference and limits your potential loss to no more than 50%. As buy arguments go, it's not the strongest -- but it's a start.

Apache chopped
In less optimistic analyst news, the folks at RBC Capital just cut $14 off their target price for Apache, now targeted to hit $112 within a year. For a stock that's pegged for 10% long-term profit growth but costs a mere eight times earnings, that seems awfully conservative.

Turns out, the problem here may lie in the "E" that makes up half of Apache's P/E equation. Sure, under GAAP accounting standards, Apache looks profitable -- cheap, even. But according to management, Apache intends to spend as much as $10 billion on capital investment this year. That's more than twice what Apache had to spend to keep the oil flowing in 2010. It's more than twice the money the company generated in the form of cash flow last year. And it suggests that Apache may not be nearly so profitable as it appears to be. Oil drilling is not a cheap business, no doubt, and you can't really blame Apache for spending what it needs to spend to get at the black gold. That said, if every penny Apache generates from selling oil has to go right back into the business, then where's the profit in all of this for investors?

Answer: There may be none.

Gentlemen, start your engines
Last but not least, we come to GM, which joined the party and also reported earnings this week – then subsequently suffered a pair of 10% price-target haircuts for its trouble. This morning, both RBC Capital Markets and UBS slashed price targets on GM -- to $36 and $30, respectively.

At first glance, this seems like utter madness. GM shares cost only five times earnings, after all. At a projected 13% long-term growth rate, that seems crazy cheap. And here's the good news: GM is in fact as cheap as it looks -- or almost so.

You see, like Apache, GM is involved in a capital-intensive business and must spend money to make money. Fortunately, though, GM doesn't have to spend quite so much as Apache does. Trailing free cash flow at this firm comes to a healthy $3.3 billion. While not quite as robust as reported income, it's still enough to make the stock look buyable at 10.4 times free cash.

While it must certainly pain investors to see Wall Street downgrade their stock on an earnings beat, the good news here is this: Wall Street is wrong. In my opinion, GM looks cheap enough to buy -- and thanks to analyst action, it even got a little cheaper today.