Occasionally, I really irritate a small group of

TSC

subscribers -- the

same

subscribers, time and time again. It happens when I reveal what one of them called in an earlier email, "your spineless unwillingness to pick

the

best; that way you have of covering your ass by naming two or three or even more stocks as the strong ones in an area. I hate that. I want to buy

the best

company, not just two or three good ones!"

Whew!

(That one went into a special file I keep of what I call Bozo Mail. You wouldn't believe how impassioned -- and, usually, how wrong-headed -- the emails in that folder are.)

But yes, I do confess that I often pick (for myself as well as for urging readers) two or three directly competitive companies to look at. Why? Why not

the

best

one

? Lily-livered? Pusillanimous? Just plain chicken?

Nope. Just smart enough to know my limits.

If it seems pretty clear to me that there are, say, two companies in an area that are both going to prosper on their own merits and also, probably, because of what's going on in their corner of the marketplace, I very often make a "straddle-buy." In other words, I buy both. Not necessarily on a 50:50 basis -- it's usually more like 60:40 or 70:30.

I do that because, while I'm usually confident to highly confident that I can pick the very short list of companies that will outperform all others in their markets, I often have much less certainty about

which

of the top two will do better.

I can stack up reason upon reason why Company A will come in first in the share-price gain game ... but, then some exogenous event -- losing or gaining a key executive, somebody's smart new product line, an especially smart or dumb move by Company B -- comes along, and my estimate of their rankings reverses. Sometimes just as fast as beach sand swirls out from under your feet when you're wading the shallows at the shore and the undertow chases a wave's energy back out to deep water.

Heck, sometimes I can't point, even with 20/20 hindsight, to exactly what flipped the results. My take: Hey, I just called the wrong one.

That's happened to me so often that I'm now cynical enough that I halfway

expect

to call the wrong one in one of these two-company races.

Simple answer: buy 'em both.

If I want to put $100K into, say, Internet infrastructure companies devoted to speeding up performance -- where I like both

Akamai

(AKAM) - Get Report

and

Inktomi

(INKT)

-- I'm much more likely to divide the funds, putting, say $60K in Inktomi and $40K in Akamai, than I am to put it all in Inktomi.

(As it happens, I like Inktomi's more traditional but high-performance caching technology rather than Akamai's more complex "content management" approach. And as explosive as Akamai has been lately, I think Inktomi is going to run faster for a while than the much marked-up Akamai. That's why I say 60% in Inktomi. But the bet would work the other way, as well.)

Gaming this, what have I done? Have I sub-optimized my likely returns? Yes, sure, some of the time. I did that by buying some insurance for the bet, straddling the two lead horses in the race. I like insurance.

But, do I get sub-optimized results overall?

No way. Because if I'm right in the first place about investing in either of the two companies (in this example, in the Net performance-boosting leaders), my two-horse parlays will produce great results. Sure, one will inevitably do better than the other. But by owning both, I get an averaged result ... usually, a very good one.

(Please note that I am not recommending either stock.

I am not long either one

, and I don't expect to be soon. This is just an example, with two particularly appropriate plays.)

I like this two-horse bet approach a lot, and use it for a substantial part of my equity investments. Some other good examples?

Say I want to put some money into computer servers.

Sun Microsystems

(SUNW) - Get Report

makes the best, most powerful Unix servers -- where "best" and "most powerful" are

always

synonymous! -- and I think

Dell

(DELL) - Get Report

is the best play on somewhat less powerful, but vastly cheaper, Intel x86-based servers to run Windows 2000 and Linux.

Which one do I buy? How do I think the near-term split between the tradition/experience/brute-force school behind big Unix servers, and the economy/simplicity/leanness arguments of Windows and Linux believers is going to break out?

Excuse me: Who cares? This isn't theology; it's investing.

Splitting my investment between Sun and Dell lets me play both. And win, probably, on both -- without worrying which edges the other over the rest of the year.

Another? How about

Lucent

(LU)

and

Nortel

(NT)

? Like 'em both. I believe Nortel looks better very short term, but Lucent probably has the mid- to long-term edge. Both will prosper, but which one will prosper the most?

I like the idea of buying both, another two-horse straddle. Since I think Nortel will have the edge for the next several months, maybe a little longer, I might go 60:40 or 70:30 Nortel, with a few bucks less on Lucent. I can always adjust that.

(Since my examples sound like micro-level sector investing, I should say that I don't particularly follow the sector approach; it's just a part of my arsenal.)

I like to make money. I like to sleep well at night. I like two-horse straddles.

Disagree? Bad plan? So send me a nasty email. But you're going to find me awfully hard to convince on this one. It may not be very elegant, and it sure isn't very technical or scientific.

But it

is

a heck of a way to make money.

Jim Seymour is president of Seymour Group, an information-strategies consulting firm working with corporate clients in the U.S., Europe and Asia, and a longtime columnist for PC Magazine. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. At time of publication, Seymour was long Lucent, although holdings can change at any time. Seymour does not write about companies that are current or recent consulting clients of Seymour Group. While Seymour cannot provide investment advice or recommendations, he invites your feedback at

jseymour@thestreet.com.