By Jack Hough

Facebook's initial public offering is slated for May 18, the Wall Street Journal reported Thursday. Between now and then, management will go on its "roadshow" to explain its strategy to potential investors, and excitement over the deal is likely to grow.

The statistics say not to buy in. Most IPOs lose money, studies show, which makes sense, because they represent a highly informed class of investors deciding to sell.

Valuation says stay away, too. Facebook brought in revenues of just over $1 billion last quarter, up 45% from the same quarter a year ago. If that growth rate holds for the rest of the year, Facebook will bring in $5.4 billion in revenues this year. The IPO details suggest a valuation of $77 billion to $96 billion, and shares could easily open much higher than the offering price. So Facebook may sell for 10 times revenues or more by the time ordinary investors buy in, more than five times as expensive as the average stock.

Then again, the statistics say not to buy scratch-off lottery tickets, but states will nonetheless sell more than $30 billion worth of them this year.

For Facebook fans who care more about the "how" than the "why" or "why not", here are some guidelines:

Post continues below.

1. Don't buy at the open
Popular dotcom IPOs in from recent years -- LinkedIn (LNKD), Groupon (GRPN), Zynga (ZNGA) -- closed lower than they opened on their first day. One past exception is Google (GOOG), but it closed only fractionally higher and at one point during its first day traded 4% below its opening price. Also, Google used a non-traditional IPO method called a Dutch auction, which tends to cut down on frantic price swings during early trading.

2. Use a "limit" order
After waiting for the open, those who buy should consider specifying a maximum price. Feel free to set it below the opening price. If it doesn't get executed, you can always change it to a “market” order for immediate execution shortly before the close of trading.

3. Bet small
Don't be swayed by Wall Street projections for how much Facebook could be worth five or 10 years from now. Valuation math is best used to help determine which stocks are good deals based on today's results or near-term forecasts.  For example, Apple (AAPL), Google, Microsoft (MSFT) and Intel (INTC) look reasonably priced relative to recent earnings, sales and cash flow.

But projections for how much companies will earn in 2018 amount to guessing. After all, even over the past quarter, analysts have been exactly right about earnings estimates for fewer than 10% of S&P 500 companies.

Similarly, don't pay too much attention to non-financial arguments that Facebook is sure to soar in value based on concepts like the network effect. The network effect refers to goods and services that become more valuable as more people use them. Facebook is an excellent example. The key is, these goods and services become more valuable to their users, and not necessarily to their investors.

Whether investors make money depends on how much they pay. The Buffett effect, in other words, trumps the network effect.

4. Don't be shy about taking profits early
Facebook is likely to debut at more than half the stock market value of Google despite having less than one-tenth of Google's profits. That's not to say that it can't soar in value. But investors should understand that they aren't getting into their Facebook investment anywhere near the ground floor. They're buying a giant company at a high price. When it comes time to decide how soon to sell, take a cue from Facebook's current owners. They're selling now.

More from SmartMoney.com

• Facebook IPO to seek $85 billion to $95 billion value
• 5 ways the government may friend Facebook
• Is Facebook a monopoly?