(MONEY Magazine) – Michael Murphy presides over a tech mini-empire: He oversees the nation's most widely read tech investing newsletters, runs three of his own mutual funds, has seen his book (Every Investor's Guide to High-Tech Stocks & Mutual Funds) go into its third printing and is a regular on CNBC and the financial conference circuit. And he owes it all to a two-year stint in prison.

When he graduated from Harvard back in 1963, none of this seemed very likely. He was going to be a lawyer. But he dropped out of law school and began experimenting with LSD. Drug use, he says, caused a "personality breakdown" that ultimately led him to rob two banks. He got caught, and in 1966 a federal judge sentenced him to 10 years.

Murphy's life story is a tale of unexpected twists and dramatic turns: He would be paroled, land a job at American Express and get a presidential pardon. Yet for all his personal intrigue, the most surprising thing about Murphy may be this: The great tech guru was in the right place at the right time with the right idea--yet amid the greatest tech bull market of all time his stock-picking record is decidedly mixed. He has been preaching the gospel of technology stocks since 1981--when Internet promoter Henry Blodget was still in high school. He has churned out reams of reports and recommendations. He boasts an encyclopedic knowledge of tech-stock history, from old mainframe companies to hot Internet IPOs. But he still has not broken the bank.

FROM PRISON TO AMEX

Michael Murphy is clearly a smart guy. The sandy-haired 58-year-old fluently rattles off statistics about tech stock after tech stock, as if he were reading from a crib sheet. "You don't have to buy Internet start-ups and biotech development stage companies to be a tech investor," he says. "We're along enough in the revolution now that there are some dominant companies you can focus on."

We're making the drive from San Francisco to his cramped offices an hour away in the sleepy coastal town of Half Moon Bay. It is a classic Murphy contradiction: the authoritative voice of a Silicon Valley insider with a nondescript office above the Cowboy Surf Shop, two blocks from the Pacific. It's a world away from the tech apex of Sand Hill Road in Palo Alto.

The location matches his unconventional career track. After receiving an economics degree from Harvard in 1963, Murphy set out to be a lawyer. Trouble was, he hated law school. He was also chronically broke and ended up dropping out. That's when his life spiraled out of control. After using LSD, he experienced an "inability to remember what day it was, what time it was," he recalls. "I'd get these hallucinations and have to pull over to the side of the road. It was like, 'I don't have any money, the bank has money, I'll go get some money out of the bank.' It sounds stupid, but that's about the extent of the logic."

The way Murphy tells the story, he bought a handgun, a Saturday night special, and walked into a bank near his girlfriend's apartment in Delaware. He walked out with $7,000. "I don't know how I got away with it." The second time he didn't. He went to jail in 1966.

To keep busy in stir, he began tinkering with computers. He was remorseful over the robberies, eager to start a new life. He was also resourceful. A member of the high-IQ society Mensa (he was clocked at 142 back in the '60s), Murphy placed a job-wanted ad in a Mensa newsletter. Richard Holmes, then chief financial officer of American Express Investment Management, happened to see the ad. He decided to give Murphy a chance. When Murphy was paroled in 1968, AmEx hired him as a computer programmer in its San Francisco office.

There Murphy got hooked on the stock market. His job was to design computer programs to track stock portfolios and help the investment guys purchase computer gear. But he learned enough about the tech industry that when AmEx's computer analyst left for Wall Street, the company asked Murphy to take the job. "He liked to crank a lot of numbers and was a very quick study," says Phil Lamoreaux, a former American Express analyst and portfolio manager who worked with Murphy. He eventually worked his way up to vice president, sitting on AmEx's investment committee. His prison record was no handicap; with the help of high-ranking company executives, Murphy got a pardon from President Richard Nixon in 1973.

Meanwhile, he was becoming increasingly obsessed with computers. In the 1970s he was one of the first to own an Apple. In 1981, when IBM introduced its first desktop computer, he was blown away by its potential. "I was absolutely convinced this was going to be huge," he says. He also believed that technology had the potential to create vast amounts of wealth for investors. Over games of bridge, Murphy and a friend came up with the idea of starting a newsletter and money-management firm specializing in tech stocks. "Back then," he recalls, "the people I'd talk to in Silicon Valley didn't have a clue what was going on in the larger sense. And they certainly didn't have access to the investment research. We thought, there has got to be a market here for decent-quality research on those companies."

MISSING THE MARK

Two decades later, Murphy is one of the most influential and provocative voices in technology. His biweekly California Technology Stock Letter and the monthly Technology Investing have 145,000 subscribers combined, making them the sector's most widely read tipsheets. His buy-and-sell pronouncements have the power to move markets.

You'd expect a guy who's been espousing the virtues of tech stocks for 20 years to be, oh, a zillionaire, with legions of followers who have become wealthy thanks to his wisdom. The truth is that Murphy's record as a stock picker is mixed at best.

Subscribers who followed Murphy's California newsletter model portfolios since 1983 would have earned just 10.3% annually, according to newsletter tracker Hulbert Financial Digest. The Wilshire 5000 gained an average of 16.9% annually during that period. Followers of Technology Investing, which began appearing in January 1999, have fared much better. The monthly's portfolio returned 157% in 1999, making it Hulbert's top-ranked newsletter for the year.

Murphy's three small mutual funds show similar mixed results. Over the past three years (through Sept. 15), the Nasdaq composite index has posted average annual returns of 36%. In contrast, the Monterey Murphy New World Technology fund posted a three-year average annual return of 2.2% and the Monterey Murphy New World Biotech fund had a return of 18.5%. His Technology Convertibles fund, which buys bonds issued by tech companies, has an average three-year return of 10.9%, ranking it 28th among the 62 funds in its peer group.

WARNING: CONTRARIAN AT WORK

Why aren't his results better? It's not for lack of effort. Like many analysts, he gets ideas by attending industry conferences and visiting companies. But he goes well beyond those basics. Two months ago, he was studying disassembled hand-held devices from Palm, Compaq and Hewlett-Packard to help him determine whether he should recommend the companies' stock. "This guy went through an incredible due-diligence process with us, right down to reading our manuals and even reading our patent," says Christopher Keene, CEO of tiny Persistence Software, which Murphy recommends in one of his model portfolios.

And his investing strategy seems sensible enough. Murphy lays out four simple criteria for picking tech stocks: annual sales growth of at least 15%, pretax profit margins of 15% or better, return on equity of 15% or higher and annual research-and-development spending of at least 7% of sales. Murphy argues that R&D is crucial because that's what generates new products and, in turn, dominant market share, high margins and high growth rates.

Perhaps his middling performance has something to do with his famous contrarian streak and his penchant for controversial, sometimes kooky pronouncements. "There's no wishy-washy Charlie Brown about Michael," says one of his former bosses, venture capitalist Reid Dennis. "He's very bright, and he's got lots of energy, lots of ideas," echoes his former partner Jim McCamant. "The problem is, that sometimes translates into doing silly things that he hasn't thought about long enough."

In 1994, Murphy embraced a renegade scientist's research study that claimed that HIV doesn't cause AIDS. As a result, he recommended that investors sell several pharmaceutical stocks, including Glaxo Wellcome and Agouron. That proved to be poor investment advice. Anyone who followed Murphy's call missed out on those stocks' huge gains. "That's his contrarian streak at work," says McCamant.

When it comes to picking computer stocks, he also goes against the grain. In the late '90s, industry giants like Cisco, Intel and Microsoft fueled the stock market's spectacular surge. But for his model portfolios and mutual funds, Murphy favored small, lesser-known names that he considered undervalued. This bias hurt his returns badly. In 1998, for example, Monterey Murphy New World Technology fund had the worst three-year performance of 44 technology mutual funds tracked by Morningstar. That dismal showing prompted one small concession: Last year Murphy added five big stocks he believes should be core holdings--Applied Materials, Cisco, Intel, Microsoft and Oracle--to his newsletter portfolio and his funds. His Technology Investing letter does focus on big stocks, which probably explains why it racked up a triple-digit return last year. Still, he is adamant about not loading up on stocks with high price/earnings ratios. "We're not going to have an average P/E of 120," he says, "because I've been through too many bear markets, and I know what's going to happen." His technology fund, he says, has an average P/E of about 30.

Another explanation for Murphy's lackluster results is that he's a market timer, constantly shifting from stocks to cash in hopes of exploiting seasonal patterns he believes he's spotted. In April, after the Nasdaq crashed, Murphy told investors to get fully invested, theorizing that the battered market would skip the traditional summer slump in tech stocks. In July he reversed course, telling subscribers he was selling all his stocks to avoid an imminent 25% drop in the Nasdaq. Didn't happen. Murphy's about-face prompted a wave of angry calls and e-mail from subscribers. "Customer service," he recalls, "came to me and said, 'Can you put something on the hotline about what a moron you are? Because a lot of people are calling and saying you're a moron.'" Murphy says he's considering halting the practice of giving his subscribers market-timing advice.

But he seems undaunted by his undistinguished performance record. He still approaches the sector with the passion of an evangelist. His mission: to demystify technology by explaining in layman's terms how it works. He's constantly telling investors that tech isn't as risky as they think and that they should put lots of their money into the sector. How much? In a risky twist on the old asset-allocation rule of thumb, Murphy recommends that you subtract your age from 100 to determine the percentage of your assets that should be in tech stocks. "I really believe the best thing you can do for people right now is get them out of the Old Economy and into the New Economy," he says.

To his mind, his most important contribution is to reach out to investors who remain afraid of tech stocks and show them they can--and must--jump in. "There's a huge number of people who haven't even started, and they really need to," he says. The question is: When it comes to choosing stocks, how much should they listen to Michael Murphy?