(FORTUNE Magazine) – A FEW MONTHS AGO, before the deluge, it seemed like a bit of stock market arcana of interest only to managers of vast sums. Now it is a prime topic of conversation, especially among people eager to fix blame for the day the Dow lost 508 points. Much of that talk has focused on Hayne Leland, John O'Brien, and Mark Rubinstein, the inventors and missionaries of portfolio insurance, a hedging technique that many think made the plunge steeper and deeper than it would have been otherwise. By 1987, six years after the three men started a company, called LOR for short, to market their method, fund managers had bought coverage for upwards of $60 billion of assets from LOR and companies under license to it. In those six years, the three say that they had only one dissatisfied major customer -- until October 19. On that day the technique failed to do what some of LOR's clients expected. One complained that his company lost more in a day than it was supposed to in three years. Several suspended their policies. Overall, only about half of that $60 billion remains covered. The recriminations reminded users and spectators alike that they had a lot to learn about what portfolio insurance really is, and what can go wrong when the roof falls in. One thing it really is not is insurance, at least in the conventional sense of a policy that guarantees a certain benefit in case of a covered loss. Instead it is an investment strategy that a client pays a financial management company to carry out. The aim is to make sure that stock portfolios decline by no more than a specified amount over a fixed period -- say, 5% over three years. Originally, portfolio insurance essentially meant hedging a portfolio by adjusting the ratio of stocks to money market instruments as stock prices rise and fall. In a simplified example, if stocks sink 8% in a bad stretch, the insurer switches some of the portfolio to money market instruments, which begin earning interest. He switches enough so that the interest earned will bring the client's assets back up to at least -5% by the end of the contract. When stock prices rise, enlarging the cushion between the new value of the portfolio and the acceptable minimum, the insurer uses cash to buy more equities. That is how portfolio insurance worked when Leland, O'Brien, and Rubinstein began selling it. Since 1984, however, they have hedged by selling Standard & Poor's 500-stock index futures -- that is, agreeing to sell a basket of the stocks in the S&P index at the current price on a specified future date. This contract locks in the present value of the portfolio if stocks fall. Index futures helped portfolio insurance take off, partly because they are cheaper, easier, and quicker to buy and sell than stocks are. HAYNE LELAND, now 46, got the germ of the idea in the mid-1970s, when his brother, an investment manager, lamented that institutional investors had been scared out of the market by the 1974 slump, only to miss out on the next year's recovery. Born in Boston, he moved at age 5 to the state of Washington, where his father went into the lumber business. Leland returned East to school, earning a bachelor's degree and a doctorate in economics from Harvard, with a master's from the London School of Economics in between. ''I was intrigued by the role of uncertainty in markets,'' he says. He has remained in academia, teaching economics at Stanford and later becoming a finance professor at the University of California at Berkeley, a post that still occupies the bulk of his time. Once Leland got the idea of portfolio insurance, he took his theory to a Berkeley colleague, Mark Rubinstein, to turn it into a product that could be marketed to big pension funds. Leland and Rubinstein had been crossing paths since they were kids without really getting to know each other -- first in Seattle, where they attended the Lakeside School (also alma mater of software impresario Bill Gates), and later at Harvard. The two men complement each other well; they even finish each other's sentences. Leland is a vigorous promoter of the company and its innovative product; Rubinstein, 43, more modest and self-deprecating, displays an academic's curiosity. In recounting the history of the company, he talks with as much enthusiasm about what slowed them down as he does of what made things finally come together. The jelling ingredient turned out to be John O'Brien. For when the two professors finally ventured off campus and tried to sell their product, nothing happened. ''We didn't really have the time and talent to be serious marketers,'' says Leland. ''I made visits to four or five banks. Then I waited for them to call me.'' Finally, in 1980, they made their first big convert -- O'Brien. A native of the Bronx, O'Brien, 50, followed a longer path to the world of finance than Leland and Rubinstein. Like his two partners, he went to college in Cambridge, but down the river at MIT, where he earned a dual degree in engineering and economics. He spent four years in the Air Force and five with the Planning Research Corp., an offshoot of the Rand Corp., whose main projects involved war-gaming and defenses against ballistic missiles. He later switched from military risks to investment risks and crossed over to the financial side for good when he was assigned to evaluate Planning Research's pension fund. EVENTUALLY O'Brien went to work for A.G. Becker, where he analyzed pension fund investments. That led him to visit Berkeley, where he saw a presentation by Leland and Rubinstein. Impressed with the commercial potential of the idea, O'Brien took it back to Becker, which rejected it as too complicated. So he quit his lucrative haven to become chief executive of Leland O'Brien & Rubinstein in 1981, bringing what the venture sorely needed: the ability to break down complicated ideas into graspable pieces. His easygoing, good- natured manner helped too. What, then, happened on October 19? Theories conflict, and different things happened to different portfolio insurers. Though no one suggests that portfolio insurance was the sole or even the principal cause of the crash, it may have contributed to the depth and rapidity of the decline. In the wake of the previous week's steep losses, portfolio insurers sold an unusually large number of futures contracts when the Chicago futures exchanges opened on Black Monday. That helped drive futures prices down sharply. Stock prices then followed suit. So portfolio insurers, and others, sold still more futures contracts, reinforcing the downward spiral. NOT EVERY INSURER was selling futures like crazy, though. Some, including LOR, stopped selling them because a gap appeared between the futures market and the stock market, with stocks trading higher. If the gap reflected a real divergence between the two markets, then selling futures contracts would lock in a value that was too low -- significantly lower than what stocks were trading for at the same time. But there was no way of knowing how much of the gap was real and how much reflected a greater time lag in recording trades on the stock exchange than on the futures exchange. In retrospect, insurers who stopped selling futures recognize that they should have continued. Messrs. Leland, O'Brien, and Rubinstein admit to being chastened by events; they are now busy figuring out what to do next. They insist that the theory underlying portfolio insurance is sound, and they want to make it perform better when markets move quickly. They are also working on a new way of hedging, to be unveiled early in 1988, that O'Brien says won't necessitate going to the stock, futures, or options markets. O'Brien's reaction has been philosophical. ''The events of October 19 didn't reduce people's concern with risk,'' he says wryly. ''In every big event there are the seeds of a new order.'' Perhaps he has been looking out his office window for inspiration -- or for a less tangible form of insurance. In big, red neon letters, a sign atop a nearby building says JESUS SAVES.