A Couple More Glimmers

12/03/08 9:10 AM EST

Good Wednesday morning...we begin the day with selling pressure in the pre-market futures as we continue to swing from buying to selling extremes.  Commodities are weak, as are most overseas equity markets and non-Yen currencies.

Yesterday we touched on the horrid breadth within the S&P 500 and the NYSE as a whole on Monday.  Only 3% of total volume flowed into the few stocks that managed to claw higher during the day, one of the worst ratios in history.

Tuesday's recovery displayed nearly the complete opposite, something we should be getting accustomed to by now.  The market has no memory from one day to the next.

Over the history of the S&P, there have been 12 times (not including the past month) when the Up Volume Ratio swung from less than 10% one day to nearly 90% the next.  They were generally successful buy signals, with the index showing positive returns over the next three months 11 of the 12 times, by an average of an impressive +6.2%.  Risk was relatively muted (the most that the S&P declined from the buy signal averaged only -2.9%), while maximum gains more than doubled that at +8.7%.

That sounds great, however there's a "but"...the last signal, generated last month on October 28th, was a miserable failure.  The S&P dropped more than 20% since that buy signal, the worst drawdown of any other historical signal by a factor of three.

That's what makes this juncture incredibly difficult.  We had so many of these types of impressively consistent buy signals in October that it seemed highly likely we would be enjoying at least an intermediate-term rally throughout November and December.  But the first three weeks of November violated all kinds of those historical precedents, throwing us into one of the most difficult markets in 110 years.

That confusion has seemingly caused newsletter writers to fully embrace the idea of a bear market, one in which all rallies should be sold.

The latest survey of newsletters from Investor's Intelligence showed a surprisingly large drop in bullish opinion despite the big recovery in stocks during most of the survey week.  The Bull Ratio (bulls / (bulls + bears)) dropped to under 32% from just under 40% the prior week.

Never in the survey's 40-year history have we seen the Bull Ratio drop by any amount on a week when the S&P rose by more than 6.3%.  The only period that comes even remotely close to what we saw last week would be the week ended May 2, 1997 when the Bull Ratio declined by 8.7% to 48% on a week when the S&P rose +6.2%.  After that, the S&P went on to six straight weeks of gains, but it was obviously under dramatically different circumstances.

Overall, there were 10 weeks when the Bull Ratio dropped by 3% or more during a week when the S&P gained 3% or more since 1969.  The following week, the S&P was up 8 of the 10 times by an average of +1.3%.  A month later it was still up 8 times but the average return climbed to +2.8%, and three months later 7 times with an average of +4.2%.

There were three other weeks when this occurred following a 52-week low in the S&P.  Those were 08/31/90, 04/13/01 and 10/11/02.  Those were within the proximity of intermediate-term lows but it's difficult to read anything consistent into the three.

Despite the November violation of all kinds of bullish precedents, there remain some glimmers of hope for an intermediate-term rally.  There are the two developments from above, and in addition the Dumb Money Confidence recently registered only 13%, the kind of level that has preceded other bear-market rallies.  Corporate insiders have been overwhelmingly buying their stock, to one of the largest degrees in history.  There is ample cash on the sidelines, enough to buy more than 36% of the entire S&P 500 just in money market funds alone (though a big chunk of that is institutional).  And early last week, equities were able to add to their gains despite a number of short-term overbought readings - the first time since July we'd seen such a thing.

Because of the November failure, though, I'm not as willing to hold intermediate-term positions than I was in October.  We simply should not have seen that kind of behavior - even during the worst markets in history, the indices had been able to piece together multi-month rallies from the kinds of extremes we witnessed two months ago.

So I want to see some more evidence that buyers are willing to put up with the uncertainty and volatility.  A great first step would be a move back over 900, helping to form an initial series of higher lows and higher highs in the S&P.

For the short-term, it's generally not a good sign to see a big down day then an "inside day" recovery.  We saw that on September 30th and November 7th, both of which turned out badly for those betting on a continued recovery.  The 850 area on the S&P remains an obvious resistance area, a level that stopped the S&P in its tracks yesterday.  Dropping below Monday's low around 815 should usher in all kinds of expectations for a test of November's low around 750.

For me, I'm not doing much in this zone, with the lack of short-term extremes we're seeing right now.  I'll likely be interested in longs if we see a spike down towards the November lows that generates another round of pessimistic extremes, or a move back over 900 (after an initial wave of likely overbought readings wear off).  From the short side, weak rallies that are unable to make it over 850 remain ripe for fading.