You never know what you're made of until faced with a defining moment of stress. Even though our combat is confined to the markets, an important parallel between trading and warfare holds: Heroes are ordinary people who do extraordinary things.

An odd characteristic of large-scale industrial warfare was the unintended development -- planned by neither army and joined at a nonstrategic site -- of a cataclysmic bloodbath won by the heretofore-losing side. The list includes

Gettysburg

,

Verdun

, and

Stalingrad

. A common thread in all of these cases was overextension by too-confident winners and a subsequent refusal to acknowledge their predicaments until it was too late to do anything about it.

You only know you're overextended after the fact. The winning side over the past few years, no question about it, has been the bulls. We now know, too late, the March highs represented a bridge too far. The trick will be acknowledging our predicament and organizing a strategic retreat.

Hey, Watch It With That Thermometer!

Volatility is the market's price for uncertainty, a sort of thermometer for risk. It is normal for stock market volatility to increase as prices fall; few investors are naturally short stock, and demand for insurance jumps as stock prices fall.

Movin' on Up
Nasdaq volatility far outpaced the S&P's measure.

Source: Bloomberg

Despite that historical trend, though,

Nasdaq

volatility was increasing during its incredible rally from October 1999 through March 2000. This unusual combination was due, most likely, to dealer and market-maker buying to maintain competitive position and to cover previous short positions.

TheStreet Recommends

Against the Grain
Proving conventional wisdom wrong, the Nasdaq enjoyed a bull market in the face of rising volatility in 1999.

Source: Bloomberg

Moreover, the dollar value of each percentage move was increasing with the market's level, and this made the market-making game an expensive proposition. These observations led to the conclusion that a market top would be reached when volatility rose during a rally: The last doubters would now be long.

We can look at the combination of price and volatility through the prism of anchoring. Anchoring is a fancy psychological term for getting used to a state of affairs and making judgments not based on any absolute standard -- such as comparing P/E ratios to anything resembling reality -- but instead based on relative standards -- such as where our personal equity was but a few fond weeks ago.

Let's be honest with ourselves: The closing level on Friday, April 14, 2000, was not breached on an intraday basis until Dec. 3, 1999, and that level seemed pretty high at the time. Had we not had the intervening 50% rally in three months, none of us would regard the present level as shockingly low. However, we anchored ourselves into a mental state of permanent rallies.

Charted Territory
The April 14 Nasdaq slide marked a major retracing of ground previously won by investors.

Source: Bloomberg

We can reorganize the data in the graph above to a scale based on gain

retracement. Each price point on the

Nasdaq 100 index

chart can be represented as a percentage loss from the all-time high; a new high will be on the left side of the scale as 0%. The largest gain retracement, by far, is the one on April 14, 2000 (this piece is being written over the weekend, and does not incorporate Monday's data).

Each volatility point can be represented as a premium or discount to the previous month's average volatility. The largest volatility premiums occurred on the market crash date of Oct. 27, 1997, the 1998 spike low date of Oct. 8, and the infamous April 14, 2000. The lowest volatility points occur at the end of October 1998; these coincide with the start of the huge rally in the NDX from 1129 on Oct. 8, 1998, to 4705 on March 10, 2000.

The relative increase in volatility increases as a function of the square of gain retracement, as highlighted by the red trend curve. This isn't surprising at all; all but the most juvenile accept some sort of retracement as part of the investing process, and we had become inured to the NDX's regular 10% corrections, three of which came this year prior to the current collapse. It's only when the truly unexpected occurs that we panic, mourn our losses, and push volatility skyward.

So, What'd the Thermometer Say?

The falling volatility/rising price combination did, in fact, define the recent NDX high. On March 9-10, 2000, the NDX's volatilities of 47.31% and 47.92% were 1.09 and 0.49 points, respectively, below the previous month's average. The real question, of course, is has it been prudent for investors to be heroes during the past two sessions?

If volatility keeps declining as it has during the past two days, stop selling and start buying. If not, a retreat may be the better part of valor.

Howard L. Simons is a professor of finance at the Illinois Institute of Technology, a trading consultant and the author of The Dynamic Option Selection System (John Wiley & Sons, 1999). Under no circumstances does the information in this column represent a recommendation to buy or sell securities. While Simons cannot provide investment advice or recommendations, he invites your feedback at

HSimons@aol.com.