Expecting The Unexpected

09/23/08 9:05 AM EST

Good Tuesday morning...We begin the day with what looks like relative calm for a change.

And what a change it would be - the S&P 500 has had an intraday range of more than 3.5% every day for the past six days, which has never happened before going back to 1982 when reliable intraday data begins.  Mid-October 1982 and mid-October 1987 both had five straight such days.

These times are just so unusual.  We're seeing price action that can only be compared to two time periods in the past 80 years, immediately after the 1987 crash and during the mid-1930's (partly coinciding with a chart we looked at yesterday regarding a previous ban on short sales).

The crazy moves aren't just in stocks.  One of the funds I track holds senior secured floating-rate loans.  These are short-term loans made to less-than-ideal corporate borrowers, but which historically present a very low default rate - first because they are short-term loans, and second because they are "senior" to most other debt.  It has almost no exposure to financials or homebuilders.

The fund, Eaton Vance Senior Income Trust (EVF), traded at a 25% discount to its Net Asset Value last week, an unheard-of haircut to current market prices in the 10 years since the fund's inception.  Previously, a discount of 15% was about all it ever saw before rebounding.

It's these types of almost-never-before-seen events that make the current environment so tricky.  I would normally be buying EVF with such a huge discount, but the fact that it just traded at a record discount raises the question "What don't I know?".

That's the problem we're facing here - a tremendous amount of uncertainty.  Traders hate uncertainty...they at least like to think they know all the variables.  That's why we're seeing such a scramble to restore confidence above all else.  Once confidence returns, then we can work through the details.

The level of uncertainty has been amply reflected in the daily price volatility and volume figures.  As I noted yesterday, the historical volatility on the S&P last week suggested that the index would move 80% one way or the other during the course of the next year, which of course it almost certainly will not.  And volume across the exchanges has been setting all-time records.

I don't normally put much weight on volume, at least in terms of the concept of "accumulation" and "distribution" days, technical theories that I think are bunk when looking at the broader market.

But there are times, when total stock volume makes one extreme or another, that it can prove to be consistently useful.  One of the remarkable aspects of trading yesterday was the drop-off in volume, making it one of the lowest-volume, large-percentage drops in the past 40 years.  Of course, any normal level of volume would look pathetic next to last week's numbers, which were record-setting.

Still, that may say something in and of itself.  The fact that we dropped nearly 4% yesterday in the S&P 500, but volume declined more than 40% from Friday's numbers is perhaps notable.

The chart below shows the correlation between large (-2% or greater) daily drops in the S&P and the daily change in total NYSE composite volume.

Since 1965, volume has increased 16% on average from the prior day when the S&P drops more than -2%.  It has jumped an average of 23% when the S&P's decline was worse than -3%.  Yesterday's decline of 42% in volume was, by far, the largest pullback in volume for a decline of more than -3% in the S&P.

There were only three other days that saw the S&P drop more than -3.5%, and volume more than 10% from the previous day.  Those were 10/25/82 (in the months following the August 1982 bear-market low), 10/22/87 (in the aftermath of the '87 crash) and 03/24/03 (in the days following the end of the last bear market).

We normally spend a lot of time looking at historical comparisons to our current situation, in terms of price behavior or the multitude of sentiment measures we watch.  When there are no precedents, which is exceptionally rare, that makes the job a whole lot tougher.

Yesterday we took a look at past attempts to ban short-selling, which have always been unsuccessful.  But we have seen several signs that compare very favorably with past intermediate-term (or longer) lows in stocks and credit (such as the spike in the Panic Button indicator, among others).

A couple of weeks ago, we looked at several studies that suggested we should see one to three weeks of extreme volatility with a downward bias (check), followed by a rabid short-term squeeze higher (check), followed by a re-test of the low (in progress?), and which ultimately leads to good fourth-quarter gains (remains to be seen).

Despite the bevy of never-before-seen developments to which we've been subjected, the market is still following that game plan pretty closely.  We're now on the back half of that forecast, the re-test of the panic low.

In 1914, the Dow crashed, bounced strongly, then spent a couple of weeks testing the low (barely exceeding it) before morphing into a bull market.

In 1929, we crashed, bounced strongly, then re-tested the low (once again exceeding it for a couple of days) before morphing into a multi-month rally.

In 1987 we crashed, bounced strongly, then spent several weeks re-testing the low (never exceeding it) before morphing into a new bull market.

That's a pretty clear pattern, and it fits with most everything else we've looked at.  Whether we've had an actual "crash" or not is debatable - we were on the precipice on Thursday morning, and we certainly did have one in the credit market, which I think is the more appropriate focus here.  So I'm sticking with my game plan, which is to continue to expect extreme volatility heading into early October, with a mostly downside bias again, before what should be an excellent fourth quarter.

I don't think there's any rush to establish intermediate-term long positions, though they should still pan out if one just buys and walks away with a very loose sell stop.  For short-term traders, it becomes more difficult with the extreme volatility, but I think the market will generally conform to overbought and oversold conditions moving forward, and I'm waiting until we see one or the other to step back into the market.