The Disappointing Nature Of Today

09/09/08 3:20 PM EST

As of: SPX 1251 HELP  ARCHIVE

This morning we went over the recent three-day pattern in the S&P 500, which according to classic technical analysis theory has a tendency to mark a bottom (a large decline, followed by another big decline with an intraday upside reversal, then a large rally day).

When we went over the nine other occurrences in the past 58 years when we've seen this kind of thing, a tendency did emerge...but it wasn't necessarily that of an imminent bottom.

What we saw consistently when looking at the precedents was that how the market reacted in the very short-term after these multi-day reversals was usually a good predictor as to how it performed going forward.  If traders immediately sold after a big rally like yesterday, then it most often lead to even more severe selling over the coming week(s).

To be more precise, out of the 9 multi-day reversal patterns we looked at, the S&P declined the following day 5 times.  The week following those, the index showed a positive return only once, with the losers averaging a dismal -3.8% (the lone exception was this past March).

When the S&P rose the next day, then its one-week forward return was positive 3 out of 4 times and averaged a healthy +4.7% (the lone exception was September 1974 when the S&P gained a bit the next day, then collapsed).

I never like to rely too heavily on just one day's trading activity, but it's safe to say that how the market reacts in the short-term (say one to three days) after reversals like we've seen have had a good history at telling us whether we've seen a major low or not.  So far, what we've seen here does not look good.

It's possible that we're now experiencing seeing something like the March low, and tonight we'll get some kind of positive resolution from Lehman (and Washington Mutual, and AIG, and, and, and...), and we'll be on our merry way to another intermediate-term rally.  Given that we never reached any kind of measureable sentiment extreme this time - unlike what we witnesses last August and November or this January, March or July - then I think that's less than likely.

More probable is that we're going to be in for a rough, and extremely volatile, couple of weeks as the fallout from the Fannie/Freddie fiasco gets sorted out and the bodies begin to float to the surface.  We always see failures among leveraged players when events like this take place, and they should come to the fore over the next several weeks.  That would fit with the pattern from above, and also the decades-long seasonal pattern in stocks.

As we continue to sell off, there will be more and more indicators and stats lining up suggesting a relief bounce.  I'm not seeing too many at the moment, other than perhaps some extremes in the Cumulative TICK and a smattering of other short-term gauges.  We're also seeing the Nasdaq 100 down for a seventh straight day.  Checking over that index's history, there were 7 other such streaks, and it managed to bounce over the next week 5 times by an average of +2.9%.  For those curious, the dates were 01/30/90, 08/03/90, 03/29/94, 11/12/97, 12/20/00, 05/17/06 and 06/12/06.

I still don't see a solid enough edge with risk to put any capital at risk in this treacherous tape which is subject to major headline risk both ways as rumors of failures, bailouts and takeovers really start to fly.  I may not unless we get finally get some hints of panic which seem well-deserved at the moment.  If we turn and rally here, it had better be soon, and I'm not so sure it won't just set up a short trade.

Notable Precedents For The Recent Reversals

09/09/08 10:20 AM EST

Good Tuesday morning...We begin the day with choppy trading in the major indices, and once again there is divergent trading among the broader sectors.  Oil is trading down and foreign markets were mostly modestly positive (ex Asia).

The performance of the S&P 500 over the past three days has certainly been extraordinary.  A greater than 2% loss on Thursday, a 1%+ loss during the day on Friday that pushed the index to at least a one-month low before a recovery to close more than 1% off the low, then a 2% rally yesterday.

That certainly seems to have the makings of a bottom, so let's check the history of the index to see if it's ever happened before.  Indeed it has, nine other times.  I'm having some major technical issues this morning, so I can't show charts of the instances, but there was an interesting pattern among the dates:

*  05/31/62 - the S&P reversed immediately and lost 12% over the next few weeks, after which it put in a bear-market bottom.

*  09/05/74 - the S&P reversed immediately and lost about 13% over the next month, after which it put in a bear-market bottom.

*  10/21/87 - the S&P reversed immediately and lost 12% over the next few days, after which it bounced for a few days, then failed again and struggled for a month before putting in a bear-market bottom.

*  12/07/87 - the S&P took off immediately and never looked back, giving excellent gains over the short- and intermediate-term.

*  09/24/01 - the S&P took off immediately and never looked back, giving excellent gains over the short- and intermediate-term.

*  07/05/02 - the S&P reversed immediately and lost about 22% over the next few weeks, after which it put in a bear-market low.

*  10/01/02 - the S&P reversed immediately and lost about 9% over the next week or so, after which it put in a bear-market bottom.

*  10/11/02 - the S&P took off immediately and never looked back, giving excellent gains over the short- and intermediate-term.

*  03/18/08 - the S&P chopped back-and-forth over the next week, before staging a decent rally that was ultimately given back.

Overall, it seemed pretty consistent that when the market reversed downward after the three-day reversal pattern, it had a lot more work to do on the downside, at least in terms of magnitude, losing 9% - 20% over the next several weeks on average.

But when the S&P followed through on the upside in the short-term, it tended to just keep going, either putting in a nice intermediate-term low or an actual bear-market bottom.  The one instance that was kind of an outlier was the last one in March, after which the S&P pulled back sharply for a day before reversing that and chopping around.

I mentioned yesterday in the Investor's Summary on the Daily Overview page that we're seeing a number of very confusing cross-currents here.  Our sentiment- and breadth-based guides are either neutral or are at conflicting extremes, and price-wise there are patterns to which both the bulls and bears could point.

One example is yesterday's major divergence between the S&P 500 and Nasdaq 100.  That was only the fifth time in history that the S&P was up 2% or more on the same day the NDX closed lower (10/20/87, 10/07/98, 03/15/00 and 01/23/08 were the others).  The NDX was up the next day 3 of 4 times, but after that it ranged all over the place.  The S&P was up the next day all four times, averaging an astounding +3.7%, but after that, the next-month returns were negative by more than -1% three times and positive once (by only +0.08%).

I make no bones about the fact that I'm supremely confused at the moment, and am hiding in cash as a result.  I can't find an edge on any time frame that seems to provide an acceptable risk/reward ratio, and I don't see the sense in risking capital in a volatile tape when that's the case.  My thought has been that the next high-probability setup would probably be a short trade as the S&P nears resistance at 1300ish, especially if it did so right away.  But that would also mean it followed through on the upside after the reversal pattern of the past few days, which has a good history at indicating major lows.  That's why I'm confused and am not doing anything trading-wise at the moment.