A few thoughts:

1) RAIL is 83% off it's high and analysts projections (guesses) about reduced future earnings are a contributing factor. When it was at it's high 18 months ago it was because analysts were making stupid projections, were wrong and it started to fall as they changed.

2) Yes, RAIL is cyclical. All the more reason to buy when it's down. The fact that it was 83% higher 18 months ago and that earnings have proven potential is better than guesing what the potential might be in the future.

3) Buffett looks to these past perfomance records all the time in making his judgements. You have this backwards ...As Warren Buffett once said, "In the business world, the rear view mirror is always clearer than the windshield." This is support for the idea that track record is important. 100% (nothing less) of Buffetts investments have been made on track record. He looks forward only to assure himself that the track record is sustainable or will return when the market realizes the hidden value that he see. He calls this Sustainable Competitive Advantage.

4) While shrinking demand is highly likely to reduce profit margins, it does not reduce a companies smarts in how to allocate resources. These folks know how to do that with geat skill and while these metrics are hard to sustain th ability to invest capital wisely and successfully remains in good times and bad and should not be cast asside.

5) To arrive at earnings estimates analysts multiply, a demand guess x a sales guess x expense guess x tax guess x energy cost guess x competition guess x a dozen other guesses to arrive a wild guess that they change after actual earnings are released.

Go back and read more Buffett

NY Times 8/15/07

Remembering a Classic Investing Theory
By DAVID LEONHARDT
More than 70 years ago, two Columbia professors named Benjamin Graham and David L. Dodd came up with a simple investing idea that remains more influential than perhaps any other. In the wake of the stock market crash in 1929, they urged investors to focus on hard facts — like a company’s past earnings and the value of its assets — rather than trying to guess what the future would bring.

Their classic 1934 textbook, “Security Analysis,” became the bible for what is now known as value investing. Warren E. Buffett took Mr. Graham’s course at Columbia Business School in the 1950s and, after working briefly for Mr. Graham’s investment firm, set out on his own to put the theories into practice. Mr. Buffett’s billions are just one part of the professors’ giant legacy.

Another Buffett protoge speaks:

"In a recent interview (http://www.fool.com/investing/value/2007/02/22/pabrais-perspectives-on-investing.aspx) with The Motley Fool's Emil Lee, Pabrai (28% annual returns) shared that he looks for stocks that scream out to him, "Buy me!" In looking at stocks he has picked, including Freight Car America (Nasdaq: RAIL) and Universal Stainless (Nasdaq: USAP), I can say they sure weren't screaming at me. Or then again, maybe I just failed to hear it."

Pabrai had nothing to do with my valuation. The interview with Pabrai is from from February. He sold before June 30 from my understanding. The stock was trading in it's tightest range of the last ten years during that period (approx 46 to 52) going nowhere. He may have simply dedcided to allocate those resources elsewhere.

Since that time the stock has dipped another 10% (from the low end of the trading range) to a level that I perceive value. We do not know if Pabrai has jumped back in or moved on to greener pastures.

If RAIL takes 2 to 5 years to return to it's glory and there is an 83% gain to be had I will be fine and ahead of market averages. If it takes longer, or I find better value somewhere else perhaps I will only see average returns. I think time, history and Buffett and Icahn buying railroad stocks are on my side.

The valuation of RAIL is highly sensitive to the context of where you look. If you look to the last year or two, it looks screamingly undervalued. If you look at the next few quarters up through about 2009/2010, you'll see the stock as being spectacularly overvalued.
Going further along is the conclusion that really matters. Yes, the next few quarters are going to be very soft, but overall this is a great company and there are many good times ahead of it as the country needs to replace a good portion of its freight fleet in the next decade. Don't just look back, don't just look 12 months forward, take a bigger and more complete view. A short piece I wrote about RAIL is at http://collegeanalysts.com/?p=45

You state first that it is too hard to predict the future, but then base your ending analysis on future projections. I think your confusion in valuing this company is a common situation for most people, and thus the value to be had.