Down 11% on Monday, then up 28% on Tuesday -- then up another 8%, down 3%, and up 2% again as the week wound on. Last week was certainly an eventful one for shareholders of satellite communications company Intelsat (I +0.00%) -- and especially so on the days immediately following earnings.

Assuming it was earnings to blame for the volatility early in the week, though, a question arises: How can one and the same set of earnings numbers cause such widely diverging reactions -- selling off Intelsat by double digits one day, and then buying them back by double digits the next?

The numbers tell the tale.

After a bit of hesitation, investors sent Intelsat stock into orbit last week. Image source: Getty Images.

Earnings? You mean "losses," right?

Let's start with the really bad news -- the headline number. On Monday, Intelsat released Q4 2017 financials showing that the company lost $0.75 per share, going by GAAP. That was a reversal of last year's $5.56 Q4 profit and literally three times as bad as the $0.25 loss that Wall Street had told investors to expect.

Intelsat's quarterly sales of $538.1 million declined 2% year over year as well, but as Intelsat CEO Stephen Spengler noted, this number was "in line with our 2017 annual guidance."

Cash famine, followed by three years of feasting

That's the bad news, and the most likely reason Intelsat's stock sank 11% right after reporting earnings, but it's not the only news Intelsat reported. Three other numbers -- or, more precisely, ranges of numbers -- help to explain why no sooner had Intelsat stock sank than it turned around and popped right back 24 hours later.

Intelsat noted that in addition to losing money under GAAP accounting standards, it burned a fair amount of cash last year: $32.8 million in negative "free cash flow" for the year as a whole. This, however, was less cash than Intelsat consumed in 2016, and according to management, things could get even better in 2018, 2019, 2020 -- and maybe even beyond.

Guiding investors on its capital spending plans over the next three years, Intelsat said it expects to spend:

• $375 million to $425 million on capital investment in 2018.
• $425 million to $500 million in 2019.
• $375 million to $475 million in 2020.

These are the sums an investor would want to subtract from Intelsat's operating cash flow to arrive at the company's free cash flow for these next three years. And how much operating cash flow does the company expect to generate?

Unfortunately, management didn't disclose that important fact. But according to data from S&P Global Market Intelligence, Intelsat has averaged $764 million in positive operating cash flow of over the past five years. And soon, says Spengler, Intelsat will begin benefiting from "better performance and economics" as it deploys its "high-throughput satellite Intelsat Epic fleet." Assuming these new sats can help Intelsat generate at least as much cash flow over the next three years, as it generated over the past five, this would imply that Intelsat might end up with:

• $364 million in free cash flow in 2018, subtracting the midpoint of the projected capex range from average operating cash flow over the past five years.
• $301.5 million in 2019.
• $364 million again in 2020.

Granted, there's some guesswork built into these estimates, but all else being equal, $343 million or so per year in annual free cash flow over the next five years seems like a reasonable expectation -- and that would be a very nice change after two straight years of negative free cash flow. If you ask me, it's this prospect for a reversal of Intelsat's free cash flow fortunes that explains the stock's quick bounce-back last week.

Valuing Intelsat

Of course, there's still the question of whether Intelsat will generate enough positive free cash flow to justify its new and improved stock price. To answer that, let's run a quick valuation check on the stock.

At today's market capitalization of $569 million, minus $525 million cash in the bank, and plus $14.4 billion in debt, Intelsat stock serves up an enterprise value of roughly $14.4 billion. Divide $343 million in free cash flow into that figure, and you get an EV/FCF ratio of 42.2.

Is that a fair price to pay for a company just turning the corner from free cash flow negative to free cash flow positive? After taking a day to think about it, investors last week appear to have concluded "yes." But I'm not sure. Forty-two times FCF seems to me an awfully high price to pay to invest in an industry at risk of being disrupted by new small satellite operators.

Until we see if Intelsat can actually deliver the free cash flow it seems to be promising -- and grow it fast enough afterwards -- I'll be sitting out this particular stock rally.