2016 isn't even half over, but already, analysts are starting to place their bets on 2017.

Case in point: On Thursday, Canaccord Genuity issued a new report in which it predicted that the S&P 500 Index (^GSPC -0.31%) -- popularly represented as the SPDR S&P 500 ETF (SPY -0.33%) -- will hit 2,340 in 2017. The analyst's new estimate was set at an even 300 points above the index's valuation Thursday -- 2,040 -- and implies a gain of nearly 15% for stocks in general over the course of the next 19 or so months.

All of which is interesting in "gee, isn't that interesting" sense. But did Canaccord just draw that 2,340 number out of a hat? Where did the number come from, and what -- specifically -- is the analyst basing its prediction on?

Here are three things you need to know.

Thing No. 1: Canaccord is bullish on the future

Broadly speaking, Canaccord argues that "a more stable global economy, emerging currency and commodity price backdrop should allow for better EPS growth despite the macro headwinds. We would focus new money in the more 'offensive' sectors." The analyst predicts that operating earnings across the S&P 500 will grow about 7.4% in 2017, to $130 per "share."

Canaccord thinks it's appropriate to pay about 18 times operating earnings for that growth. 18 times $130 equals 2,340 -- and that's where the analyst's prediction comes from.

Image source: Wikimedia Commons.

As for the details that support that prediction, Canaccord says it is "confident there is no recession on the horizon based on (1) the history of the Real Fed Funds Rate using core inflation, (2) the six-month/10-year yield curve, and (3) the Chicago Fed National Financial Condition sub-indices that measure financial market and banking stresses." (Or translated into English, all the usual statistics that predict an impending recession are flashing green today, and there's no indication that a recession is imminent.)

Meanwhile, Canaccord is encouraged by "74 months in a row of Payroll growth, Weekly Initial Unemployment Insurance Claims are at a multi-decade low, Household Debt Service Ratios and Consumer Delinquency rates are near record lows, and Consumer Confidence is near a cycle high."

And if we were to translate that one for you, Canaccord is saying that not only does it see no red flags warning of recession, but in fact wage growth, jobs growth, and consumer confidence all suggest that consumers are ready and able to spend, and that should help keep the U.S. economy humming.

Thing No. 2: Canaccord is optimistic -- but not that bright

So that's good news, right? Well, not so fast. As it turns out, whatever Canaccord Genuity's insight into macroeconomics, the analyst hasn't proven to be all that bright about valuing stocks based on that insight. Fact is, according to our analysis of its record on Motley Fool CAPS, only about 38% of Canaccord's recommendations to buy or sell stocks actually outperform the S&P, and the analyst ranks in the bottom 20% of investors we track.

It's not hard to see why. In fact, you could guess that Canaccord is an analyst inclined to overpay for equities just from reviewing its math up above. Even assuming that the analyst is right about its predicted $130 in earnings next year, for the S&P 500 index to be worth 18 times those earnings, you'd usually want to see the stocks in question growing earnings at 18% or more.

In fact though, Canaccord's prediction is for just 7.4% growth -- half the level needed to justify a valuation of 18 times. In short, Canaccord appears to be recommending that investors overpay for growth.

Thing No. 3: So why listen to this analyst at all?

That's actually not uncommon among professional stock market analysts. Smart as these guys are, and as many degrees and credentials as they hold, our data on CAPS show that only 73 out of the 218 big-name investment bankers we track boast records of better than 50% accuracy when recommending individual stocks to buy.

That means barely one analyst in three has proven itself able to beat the market over time.

The most important thing: What it means to you

So what's the moral of this story? Tracking and crunching the numbers on big macroeconomic trends is a useful service that Canaccord provides -- just don't get too hung up on the analyst's predictions of specific price targets for the S&P 500, or for individual stocks, either. Given Canaccord's (and Wall Street in general's) record of over-optimism about how high stocks can fly, you're better off doing your own homework before actually buying or selling a given stock.

Examples? Over the past week, Canaccord has made two specific stock recommendations, initiating coverage of Microsoft (MSFT -0.47%) at hold and upgrading shares of Nokia (NOK 1.36%) to buy. But honestly, their guesses are probably just as good as yours, or mine.

Microsoft, priced at 39 times earnings and expected to grow those earnings at only 8.5%, offers a good example of Canaccord being overoptimistic -- just as it is in regard to the S&P 500 index as a whole. 8.5% growth isn't anywhere near fast enough to justify paying 39 times earnings, and you don't have to be a professional stock market analyst to know that Microsoft is almost certainly overpriced. It's clearly not a buy, and probably doesn't deserve Canaccord's hold rating, either.

On the other hand, Nokia sells for less than 11 times earnings, is projected to grow at nearly 16%, and pays a 3.5% dividend yield, to boot. Nokia stock bears all the hallmarks of an excellent value play, and is a good example of a stock being cheap enough that even Wall Street can't fail to recognize it.

When you get right down to it, Wall Street may be right or wrong on the big economic trends, but valuation still matters most to successful stockpicking. And given Wall Street's record, you're probably at least as it is at spotting a good value.