Math Whiz’s Newport Office Untouched by Indictments

Princeton/Newport Partners is the brainchild of a Newport Beach mathematician who turned a successful gambling system into a computerized investing program and a New York stock broker who was tired of buying shares for others.

The former broker, James Sutton Regan, was one of five Princeton/Newport partners indicted Thursday by a federal grand jury in New York in connection with a federal investigation of stock manipulation and tax evasion.

The Orange County math whiz, Edward Oakley Thorp, was not included in the indictments. In fact, Thorp said he had little contact with federal investigators during the two years they have been checking into the investment group’s activities.

Thorp is a former university mathematics professor who gained fame in 1962 with the publication of “Beat the Dealer,” a book that proved it is possible to win at blackjack by following a card-counting system.

That he was not touched by the grand jury investigation may stem in large part from the fact that Thorp and the number-crunching end of the business remained in Newport Beach while Regan set up the trading side of the operation in Princeton, N.J.

That decision was made when Princeton/Newport was founded in 1969--in part with $25,000 of Thorp’s blackjack winnings and $100,000 in royalties from the book that the wiry, red-headed mathematician wrote after being banned from casinos in Nevada.

Because of the bicoastal bifurcation, Thorp’s operation has little contact with Regan’s other than the continual electronic transmission of data from the computers in California to the trading desks in New Jersey. The separation apparently convinced federal investigators that Thorp and his crew of 40 were not involved.

In a telephone interview from his offices in a plush Newport Center high-rise Thursday afternoon, Thorp talked about the business he and Regan have built from a two-man operation into a significant Wall Street player specializing in hedged investments.

Princeton/Newport has provided its partners--there are now about 75--with a compounded annual return of 19.6% since its founding and has never had a money-losing quarter, Thorp said.

For the first six months of 1988, however, the partnership has earned a return of only 6.5% on its investments, Thorp said. “But that’s 13%, annualized,” he added.

Thorp said he believes that Regan and the other Princeton/Newport officials are innocent and that the government has manufactured a case to pressure them into testifying against officials at several major brokerages as part of a broadening investigation of insider trading activities on Wall Street.

The 35-count indictment alleges racketeering, fraud and conspiracy on the part of the five Princeton/Newport partners and a former bond trader with Drexel Burnham Lambert, a Wall Street securities firm. The indictment says the firms took part in phony, prearranged stock trades to give Princeton/Newport $13 million in tax benefits and to enable Drexel to manipulate stock prices and avoid regulatory requirements.

“One thing that makes the government charges so absurd,” Thorp said, “is that our investment strategies are mechanical. They don’t involve the mischief we are being charged with.”

Those investment strategies are what have set Princeton/Newport apart from other securities investment partnerships over the years.

Hedging is a relatively conservative investment strategy involving simultaneous purchases and sales of similar securities to generate income with little or no market risk.

What Thorp brought to it was a proprietary system for using computers to pick the best investment choices.

“We try to come up with mathematical models and quantitative statistical ideas for identifying misvalued securities, and then we try to take a position in those securities,” Thorp said. “We go long on undervalued securities and short on those that are overvalued, and then we hedge the investment with something that tends to cancel out our risk. We might, for instance, go long on undervalued warrants and short on the underlying stock.”

In Newport Beach, Thorp heads a team of 40, including other mathematicians, statisticians, economists, computer scientists and specialists in business and finance. Together, they develop and feed data into the computer programs they have designed. The data enables the computers to identify choice investment targets--including options, warrants, bonds and futures contracts as well as common stocks.

In 1986, Princeton/Newport said its computer operations were gobbling up more than $4 million a year in operating funds.

At the time, Princeton/Newport’s equity was $150 million. Today, Thorp said, the partnership’s net worth is close to $270 million.

Investors Restricted

The investment group is not one that welcomes all comers. Thorp recently said the minimum investment for an individual partner is $2 million, which requires a net worth of $10 million, which limits the investing pool to less than 5,000 individuals in the United States.

But just waving $2 million doesn’t make Princeton/Newport open its doors. Thorp said the partnership has gone for several years at a time without adding new members.

“We are a collection of private individuals who invest as a partnership” Thorp said. “We are all sophisticated, qualified investors.”

He said he does not believe that the indictments will threaten the stability of the partnership, largely because of its exclusive nature.

Thorp has always maintained that the kind of investing Princeton/Newport does is very conservative. That claim to conservatism is a hallmark of his career.

In the late 1950s and early 1960s, he made playing blackjack in Las Vegas casinos a conservative pastime--at least for a player with a mathematician’s mind.

Thorp, who turns 56 next week, was a math teacher at UCLA in 1958 when he picked up a copy of the Journal of the American Statistical Assn. and read an article that said keeping track of the cards dealt in a blackjack game and betting according to statistical probabilities could reduce the casino’s edge so that the players could win more frequently.

Las Vegas Experiment

In a 1982 interview, Thorp said he tried out the system during a Christmas visit to Las Vegas in 1958 and found that it worked. He also remembered something from an advanced course in probability theory that led him to believe he could refine the system and go from being an occasional winner to breaking the bank.

In 1959, Thorp took a job teaching at the Massachusetts Institute of Technology and, still intrigued by the thought of a winning blackjack system, taught himself to program a computer and began force-feeding a high-speed IBM with blackjack data he had been collecting for two years.

In 1961, using computer-generated calculations, he wrote a paper on blackjack probability and delivered it at a math conference in Washington.

The paper generated enough interest that several groups offered to finance Thorp if he would go to Las Vegas and try his theories. For several months he did just that, consistently winning at the game and finally being barred by the casinos.

His revenge was to write “Beat the Dealer,” a book that is considered the bible of serious blackjack players and that has sold well over 500,000 copies.

For the next seven years, Thorp taught math--he moved from MIT to UC Irvine when it opened in 1965--and tried to find something else to challenge his taste for adventure.

He found it in Wall Street, focusing on the basic strategy of hedging and developing a math formula and computer program to make the operation as foolproof as possible.

In 1969 he met Regan, a Dartmouth graduate with a degree in philosophy, and the two started Princeton/Newport. In 1982, Thorp resigned his teaching post to devote himself full time to the partnership.