On Thursday, defense contractor L-3 Communications (NYSE:LLL) reported its Q3 earnings results. In the days since, we've seen several articles from the AP, Business Week, and MarketWatch digesting and restating the quarterly results, describing how the stock has leapt 8% on the back of those results, and the subsequent analyst upgrade from McGraw-Hill's rating arm, S&P.

So how's about today we do something a little different? Today, let's take a look at the year in review, at what it suggests to us about the rest of the year -- and why that suggestion can be misleading.

The year in review
As of last week, we're three quarters of the way through L-3's fiscal year. By now, at least in theory, we should have enough data stored up to predict where the firm is headed for the year (in the parlance of the trade, a "run rate.") So let's give that a try. Beginning with the basics, so far this year, L-3 has:

• Grown its sales 39% to \$9.1 billion vs. the first three quarters of last year.
• Seen its net income fall 1% to \$352.5 million.
• Watched earnings per diluted share fall 4% to \$2.84 per share.
• Grown free cash flow 24% to \$634 million.

To estimate the firm's run rate, you simply take the above numbers and annualize them. For example, if you have one quarter's statistics to work with, multiply them by four. If you have two quarters, multiply by two. And in today's situation, with three quarters to work with, you want to multiply the results by 1.33 to come up with a likely result for the full year. But as we'll see later, this is very tricky.

The year in anticipation
So, what the above numbers suggest is that, through the end of the current fourth quarter, L-3 is trending toward \$12.1 billion in sales, \$469 million in net profits, \$3.78 per share, and \$843 million in free cash flow. But what's interesting about those numbers is that only one of them -- sales -- bears even a passing resemblance to what both L-3 itself, and its analysts, predict (respectively, \$12.4 billion or \$12.2 billion in sales, and \$4.16 or \$4.21 per share in profits).

Why the disconnect? Ah, there, dear Fool, lies the true point of today's column: Run rates aren't always what they're cracked up to be, for three reasons.

1. Forests and trees
You know the old saw about how it's sometimes hard to see the forest for the trees? Well, it's especially hard to see the forest if one of the trees happens to fall on your head. That's pretty much what happened at L-3 back in Q2, when the firm announced a charge to earnings of \$0.83 per share. That charge, taken to account for the potential need to pay a large jury verdict and account for stock options, won't recur in Q4 (although a smaller stock options charge should be expected.) Yet if you try to work a run rate without accounting for the special charge, you're likely to greatly undershoot where L-3 will actually end up this year.

2. What if we ignore the charges?
Sorry. That won't work here, either. Back out Q2's special charges, work your calculator keys, and you'll get a run rate of \$4.88 per share in putative full-year profits. This time, you're far ahead of what either the analysts or L-3 itself predict. One reason that simply ignoring the charges doesn't work for L-3 is due to the nature of the business, which heavily tilts toward "inorganic" growth through acquisitions. So far this year, fully 30 percentage points worth of L-3's 39% sales growth came from acquisitions, such as its June 2005 purchase of Titan. Such acquisitions tend to cause "lumpy" earnings that are harder to project out using run rates.

3. One small, red fish
One final thing to consider when working a run rate is the seasonality of the business. Now, L-3's business actually isn't particularly seasonal -- on the contrary, the firm's revenues seem to just go up and up, quarter-in and quarter-out, year after year -- so this issue is a bit of a red herring in the instant case. But as long as we're going over the problems with run rates, I don't want to leave this one lying around undiscussed.

Retail is the classic example of the kind of seasonal business that defies the use of run rates. Companies like Best Buy (NYSE:BBY) or Circuit City (NYSE:CC), which do the bulk of their annual business in calendar Q4, are nigh on impossible to "run rate." No matter how well or badly they do in the first nine months of the year, Q4 has the power to make their annual results a success or failure all on its own. For example, if you tried to do a run rate for Best Buy last year based on its first three quarters' results, you'd likely have predicted annual diluted profits of \$1.31 per share -- when the actual profits were \$2.27 for the year! Similarly, event-driven (i.e., holiday) seasonality makes it difficult to work run rates on a company like 1-800-Flowers (NASDAQ:FLWS) or FTD (NYSE:FTD). There, it's the romantically bereft third calendar quarter that tends to skew run rates low.

Since we're talking about a defense contractor here, though, let's close with a defense-contracting example. As I warned first in its Foolish Forecast, and then later reemphasized in my earnings story, L-3 peer General Dynamics (NYSE:GD) has an interesting seasonality to its business as well. Although seasonality doesn't affect its revenues much at all, for some reason, the firm loads up on free cash flow in Q4, presenting investors with a quintessentially retailer-ish conundrum: The first-nine-month results may not be reflective of annual performance, at least on a cash flow basis.

In conclusion, while every story has at least one moral, this one has three. Run rates are a great tool for predicting how a steady, organic grower will perform in the future. But before putting too much faith in one, make sure to check a company's history for signs of:

• Special charges ...
• Lumpy earnings, and ...
• Seasonality.