A Weak "Oversold On Support" Setup

09/03/08 2:45 PM EST

As of: SPX 1251 HELP  ARCHIVE

Yesterday we went over the divergence between sectors, with Retail and Banks holding well but Technology and Energy getting hit hard.  That theme is evident again today, with Semiconductors just an inch away from multi-year lows and leading the Nasdaq 100 towards its third consecutive 1% decline.

This would be the fifth time we've seen three straight 1% down days in the NDX since the highs last summer (08/16/07, 11/09/07, 01/17/08 and 02/06/08).  The index bounced higher the next day three of those times, and each time it was within two days of a multi-day (at least) rally.

It's unusual that breadth on the Nasdaq is holding so well in spite of the 1% drop in the NDX, probably owing to the influence of the Semiconductor sector's weight on the NDX.  Other technology-related sectors aren't seeing nearly the level of selling pressure that Semis are.

Over the history of the NDX, there were 76 other days where it declined 1% or more yet breadth on the Nasdaq was positive.  While it seems like that should be a major positive, in reality it really wasn't - the NDX was up the next day 47% of the time and showed an average return of -0.4%.  Its performance after three and five days was almost exactly flat (about a 50% winning percentage and barely positive average return).  It has only happened once in the past five years, that being 01/16/08.  It sold off about 4% over the next few sessions before bouncing.

While the S&P 500 isn't showing the same magnitude of declines, it has still slid three straight days, including more than a 1% drop to kick things off two days ago.  We've seen this pattern 15 times since last summer's market top (those 15 instance include some overlapping days), and like the NDX study above it generally led to higher returns.  The S&P was up three days later 73% of the time by an average of +0.8%.

I mentioned yesterday and this morning that another down day should push our shortest-term guides firmly into oversold territory and perhaps set up a long-side trade against 1260ish on the S&P, which has acted as a support level recently.  The choppy rally earlier this morning helped to alleviate some of those oversold conditions, with the Price Oscillator, Cumulative TICK and VIX Strength moving back to neutral and not giving the STEM.MR Model time to really become oversold.  Perhaps that's just nitpicking, but with the "iffy" technical condition of the indices, I'd prefer to see solid oversold readings if trying to buy.

We seem to be on the precipice here, and the buyers had better step up quickly.  I could make a weak argument that we're "oversold on support" within at best a neutral intermediate-term trend, which is only a marginally bullish setup.  Frankly, I'm still not sure if I want to try to trade it, and will likely wait 'til closer to the close to decide.  Any move below 1260ish wouldn't be a good sign and in that case I would not be looking to take a long-side trade unless we either had a big selling flush or rallied and held back above 1265.

Short-term Guides Not Quite There

09/03/08 10:30 AM EST

Good Wednesday morning...We begin the new day without much movement in the broader equity indices, as Oil continues to fall and we see more follow-through on yesterday's theme of strong Banks and Retail, but weak Technology.  Historically, such sector divergences didn't lead to anything consistent for the broader averages like the S&P 500, but it's helping to keep a lid on any big directional move so far.

Yesterday we took a look at several short-term factors that suggested perhaps a little more downside follow-through, then a likely bounce in the coming days.  One of the reasons was the price pattern that indices like the S&P 500 had carved out - while it looks ugly on a chart, days like yesterday haven't necessarily been all that bearish going forward.

Another wrinkle with regards to yesterday is that breadth on the NYSE was actually positive (i.e. more stocks closed higher than their previous close than down).  If we go back and check the history of the S&P 500 tracking fund, SPY, for any time it carved out an outside down day (a higher high and lower low than the previous day, with a negative close) but yet NYSE breadth was positive, we see that there were 43 instances.  By three trading days later, it was positive 65% of the time with an average of +0.5%.

That OK, but not great.  Interestingly, though, since 2000 the pattern has occurred 23 times and the S&P showed a positive return three days later 78% of the time, averaging +0.7%, and with an average maximum reward (+1.4%) that nearly doubled the average maximum loss (-0.8%) during the three-day periods.

With our most sensitive indicators close to oversold territory, and the other patterns we looked at yesterday, it seems likely that we'll see a multi-day bounce should we get more weakness today.  Perhaps we're already in the midst of such a bounce, but I don't see the kind of setup yet that has me interested in trading it - with the questionable technical condition of the market, I'm not interested in risking capital on marginal setups.  The S&P will also be testing the lower end of its recent range around 1260ish - I'm not sure how much longer we can count on such obvious support to hold, but given the other factors it might be worth a shot if we see it today.