If stocks like

Broadcom

and

Juniper Networks

(JNPR) - Get Report

are the new leaders in the technology sector, when is a good time to buy them?

They're certainly cheaper now than they were six months ago, thanks to the bear market in technology stocks. But as the carnage on Oct. 25 following

Nortel Networks'

(NT)

earnings announcement demonstrated, cheaper stocks can still get a lot cheaper -- and very quickly. Juniper Networks, which closed at $218.13 on Oct. 24, well off its 52-week high of $244.50, quickly dropped another $15 a share, or 7%, on Oct. 25.

We'd all like to own shares of the next-generation technology leaders, but pulling the trigger is hard in a volatile market -- and it might not even be smart. That's why I decided to take a look at the question: How much does it matter that you pick the right emerging technology superstar, if you get the price wrong?

My conclusions? If you've found the next superstar technology stock before anyone else -- that is, while it's selling for a market capitalization of $3 billion or less -- you shouldn't worry about whether or not you're buying at a temporary bottom or a temporary top. The returns for catching a stock like this in its early stages are so huge that they overwhelm differences in purchase price.

But if the market has already anointed the stock as a potential leader -- and it's trading at a market cap of $30 billion or more -- then hitting a good entry price matters a great deal. In this volatile market, I'd wait to buy.

Sweet Spot in the Middle

In between the undiscovered cheap stocks and the too-obvious and too-expensive ones lies what I'd call the sweet spot of the technology sector. These are stocks that have the potential to be sector leaders but that are trading below $20 billion in market capitalization right now. At these prices, these companies don't have to hit home runs for an investor to score big. And hitting the exact bottom on these isn't critical, which gives you time to wait and see how the market develops over the next few months.

I started my thinking about this timing issue with a look back at

America Online

(AOL)

in 1993. At that time, America Online was a promising but controversial technology company struggling to take the online idea pioneered by Compuserve and Prodigy to a mass consumer market. The stock, which had gone public in March 1992, was typically volatile; after all, the company had far more concept than revenue. In the spring of 1993, for example, it traded to a split-adjusted high of $0.23 on March 12 before it began a plunge that would take it down about 25% to $0.17 on April 20.

Now, if you managed to snap up shares at that April low and hold them through the close on Oct. 24, 2000, you would have done OK for yourself. A $10,000 investment in America Online on April 20, 1993 would have been worth just over $2.7 million on Oct. 24, 2000. That's a 26,995% gain.

Your returns if you'd instead have bought at the spring top? A $10,000 initial investment on March 12, 1993 would have grown to a little more than $2 million by Oct. 24, 2000, a gain of 20,279%.

I think there are two very different, but equally valid, ways to look at those returns. First, buying at the bottom is clearly much more profitable than buying at a top. I certainly don't sneeze at the extra $700,000, or 35%, the investor who bought at the April low made in comparison to the investor who bought at the March high.

But second, that poor investor who bought at the March high wasn't, well, poor. Buying at the top was immensely more profitable than not buying at all, or buying a stock that's not even listed anymore.

Changing the potential gains from buying America Online doesn't change the financial results much, but I think it does change how we feel about those results. And the biggest change is how we feel about the cost of not buying the stock at all.

Let's say AOL didn't turn out to be one of the best investments of the last decade. Instead, assume it was just a better-than-average stock. What if AOL had climbed 438% from March 12, 1993, through the present -- exactly double the performance of the

Standard & Poor's 500

, but way below the 20,000%-plus the stock actually gained? An investment of $10,000 would grow to $53,800. The investor who caught the April bottom instead of the March 1993 top would have seen that $10,000 grow to $72,400 now instead of $53,800. That's exactly the same 35% difference that separated the first set of results. But at roughly $20,000 instead of $700,000, the results feel very different. Missing out on $700,000 because I bought at the wrong time feels much more crushing than missing out on a potential $20,000.

But it's the cost of doing nothing at all that really feels different in the two examples. If I do nothing with my $10,000 and the stock simply doubles the gains on the index, I've lost out on at least $43,800 in gains. If I do nothing and the stock turns out to be "the next AOL" I've lost out on at least $1.99 million. Don't know about you, but that gets my attention.

It's Not 1993

The key to the phenomenal results from investing in America Online, whether purchased at the temporary top or bottom, was the initial cheapness of the stock. In 1993, America Online traded at a little less than six times revenues and a market cap of about $500 million. But the technology market in 2000 is different from the technology market in 1993 in one critical way: Untried technology companies with immense potential are much, much more expensive right out of the blocks now than they were in 1993. Numbers like America Online's back then appear quaint these days.

Look down the list of current superstars. Juniper Networks sports a price-to-sales ratio of 168;

Brocade Communications

(BRCD)

, 128;

Ariba

(ARBA)

, 114;

Network Appliance

(NTAP) - Get Report

, 110. The market capitalization of Juniper Networks is already $69 billion. For Brocade, $29 billion; Ariba, $31 billion; Network Appliance, $42 billion.

Why is that important to my 1993 America Online example? Because stocks trading at multiples like this are less likely to achieve the kinds of gains racked up by America Online.

Think about it this way. To produce a 438% gain in price over seven-plus years, America Online only had to grow from a market capitalization of about $500 million to one slightly less than $2.7 billion. To produce a 438% gain from its current market capitalization of almost $70 billion, Juniper Networks would have to grow to a market capitalization of $371 billion. Only two companies on the current market --

Cisco Systems

(CSCO) - Get Report

at $386 billion and

General Electric

(GE) - Get Report

at $528 billion -- are above that size. Gaining 20,000% would result in a market capitalization for Juniper Networks of $14 trillion -- or roughly twice the current total gross domestic product of the U.S.

TST Recommends

What do all these examples and numbers add up to? I think they explain the central paradox of the current technology market:

First, it's clear that hitting the next AOL is a tremendous prize. Seeing $10,000 grow into $2.7 million, or even $2 million, is stunning.

Second, price is really unimportant if you think you've found the next AOL. Why gamble on waiting for a drop when the worst that will happen is that you're $10,000 will turn into $2 million instead of $2.7 million?

Three, the lure of that huge payoff tends to bid up the price of any young company that has the potential to reach this goal. If it doesn't matter whether you pay $210 or $220 for Juniper Networks, assuming it has the long-term potential that you want, then why not $230 or $240?

Four, that willingness to bid up potential technology superstars makes it less likely that they will produce the fabulous gains that led to the bidding frenzy in the first place. I can imagine Juniper Networks at $370 billion in market cap in seven years, although I think it unlikely. But I can't imagine the stock being worth twice as much as the entire current U.S. economy.

Those four observations suggest a strategy to me for buying the next generation of technology leaders.

A New Strategy

If you think you've found the next superstar technology stock, but so has everyone else, waiting for a deep dip may be well worth it. The odds say that each rung of success on the market capitalization ladder gets harder to climb. There are twice as many companies in this market with market caps above $20 billion as there are companies with market caps above $50 billion. I think you can increase your probability for long-term gains by waiting until a short-term problem has knocked a potential leader down from $70 billion to $50 billion, or from $53 billion to $30 billion. In other words I wouldn't be buying the Juniper Networks, the Broadcoms, the

JDS Uniphases

(JDSU)

, and the Network Appliances of this market quite yet.

In the middle of the market, among technology stocks with market capitalizations of less than $20 billion, you'll find potential new technology leaders that could drop another 25% or so if the bear takes another bite. But the stocks are also cheap enough at the moment that even if your pick doesn't turn out to be the next America Online, it still stands a good chance of doubling within a year or two.

Foundry Networks

and competitor

Extreme Neworks

(EXTR) - Get Report

are both selling for less than $10 billion in market capitalization at the moment.

ART Technology Group

(ARTG)

goes for about $6 billion, and competitor BroadVision is priced just a bit higher at $7 billion.

Research in Motion

(RIMM)

and competitor

Phone.com

(PHCM)

both have market capitalizations below $9 billion.

E*Trade Group

(EGRP)

goes for just $4.4 billion, and fellow Internet pioneer

eBay

(EBAY) - Get Report

, for $14.5 billion. Given the potential in this group, I don't think you have to buy precisely at the bottom. Give the market time to declare its direction and buy in this group when you think you see clear indications that the worst of the bear market is over.

And finally, if you've found the next superstar technology stock -- but either nobody else has, or they've decided for some reason to abandon the stock so it's really cheap -- don't worry about price. In fact, instead of spending time on charts or trying to predict the next dip, put your effort into the more important task of making sure that the company is indeed all you think it will be. For example,

Critical Path

(CPTH)

and

RealNetworks

(RNWK) - Get Report

are each selling close to $3 billion in market capitalization. If the stories behind these companies are real, then they are buys at their current prices. Or 35% higher, or 35% lower. It doesn't matter much.

In this one small part of the market, at least, you can take a vacation from obsessively watching prices.

At the time of publication, Jim Jubak owned or controlled shares in the following equities mentioned in this column: America Online, Ariba, Broadcom, Broadvison, Cisco Systems, Extreme Networks, E*Trade, GlobeSpan, Intel, LSI Logic, Microsoft, Nortel and Texas Instruments. He welcomes your feedback at

mctsc@microsoft.com.

More from

MSN MoneyCentral

Rowland's Start Investing Portfolio

Jubak's Picks

Markman's SuperModels