Overall, A Pretty Progressive Week

12/12/08 4:10 PM EST

As of: SPX 1045 HELP  ARCHIVE

There's never a dull day, much less week, in this market anymore.  Maybe it's a case of "be careful what you wish for", but it sure would be nice to have a few boring weeks again at some point.

We started the week with a bang once again, as Monday showed a large opening gap from Friday's close and most of the major indices closed up by several percent.  Over the past 20 weeks (nearly six months), only 3 of them have shown an absolute return of less than 1% on Monday - another sign of the incredible volatility to which we've been subject.

Monday's gap open pretty much marked the end of any directional movement for the next couple of days, as Tuesday's and Wednesday's trading was confined entirely within Monday's range.  The volatility contracted so much that for one of the only times in history, the S&P 500 was within a couple of hours of carving out three consecutive "inside days".

There was still some movement under the hood, though, and a few of our indicators started to show some extremes.

On Monday and Tuesday, for example, we studied a couple of different breadth measurements, and how they'd become so overbought that they were historically meaningful.  Overbought conditions during bear markets are ripe for selling...with one common exception.  When we get extremely overbought readings that follow extremely oversold ones, we get a rising market over the ensuing month(s) nearly every time.

The ability of the market to hold up after Monday's gap seemed like a positive development, though by mid-week we did have occasion to touch on a few minor negatives.  There were some bearish (for the market) readings like the OEX Put/Call Ratio and Rydex Beta Chase Index, most of the short-term indicators we looked at on December 4th were close to going "all in" on the overbought side again, and the Intermediate-term Indicator Score reached a level that has seen the market roll over every time since mid-2007.

Those were enough to get me thinking we'd have a slightly better chance at breaking down out of the congestion of the past few days rather than up, but not enough to actually trade it.  Plus, as noted on Thursday, when markets break out of a tight volatility coil like we saw mid-week, the initial move out of the congestion tends to be "false", and we get a larger and longer-lasting move in the other direction about 70% of the time.

That breakdown culminated in a huge gap down this morning as traders got hit with a plethora of bad news after the close on Thursday.  Before the open today, we looked at some stats related to prior large gaps down, and it has been a consistently good buy signal, at least in the short-term.  That proved true once again, as the S&P recovered slightly more than its historical average between today's open and close.

In the process, the major indices held the support levels we'd been looking at for awhile, and combined with everything else we discussed this week, the outlook over the coming month(s) continues to look alright.  We had a vaguely similar scenario in October, which failed miserably during early November, and that's definitely a concern.  But given the developments this week and the other positives that have been gathering, I'm willing to give the idea of an imminent intermediate-term rally another shot.

Have a safe and relaxing weekend and we'll see you next week.

Gaps Tend To Reverse, Will Test Support

12/12/08 8:50 AM EST

Good Friday morning...we begin the day with an ugly open as traders got bombarded with negative news after the close yesterday.  The worst of all is probably the Madoff fraud, as it once again shakes already-precarious investor confidence in the financial system, and could easily lead to yet more forced selling and fund collapses.

In a continuation of the tight correlations we've been seeing, overseas markets were down 4%-5%, commodities and non-Yen currencies are pulling back and bond yields are trading at either record lows or very close to it.

While it certainly looks ugly at the moment, gaps down of 3% or more in the S&P 500 have been relatively benign for the market, as it has tended to recover from the open to the close.  Out of 8 instances over the 15-year history of the S&P tracking fund (SPY), 7 of them closed the day higher than the open, averaging a return of +3.2%.  The one loser was -1.3%.

Four of them occurred this year, with three of them closing higher than the open (01/22/08, 10/10/08 and 10/24/08).  With the exception of one on September 15th, they marked short-term exhaustion points as the market rallied for at least the next couple of days.  The one in September led to a couple more days of selling pressure before a quick rally that erased the losses.

Across all 8 occurrences, the most the S&P lost during the day averaged -1.0%, compared to a maximum gain that averaged +5.9%, so a pretty skewed ratio there.  Only one of them lost more than -1.5% at any point during the day from the opening print, while every one of them gained more than +3% at some point during the day.  Three of them occurred on a Friday, and all marked short-term exhaustion points (09/21/01, 10/10/08 and 10/24/08).

If we do a slightly different test, and look for any time the S&P futures lost 2% one day then gapped down at least 2% the next morning, buying the open and holding for three to four days resulted in a winning trade each time, with one big exception - October 19, 1987 (aka "Black Monday").  The rallies averaged +4.2%, but were volatile - the average maximum loss during the trades was -4.3% (without the 1987 outlier), compared to an average max gain of +9.9%.

Over the past couple of days, we've discussed a few minor negatives.  We'd seen some bearish (for the market) readings like the OEX Put/Call Ratio and Rydex Beta Chase Index, most of the short-term indicators we looked at on December 4th were close to going "all in" on the overbought side, and the Intermediate-term Indicator Score reached a level that has seen the market roll over every time since mid-2007.

But my take has been that several of the overbought indicators we saw recently were so overbought that they were historically meaningful.  We looked at this on Monday and Tuesday, for example.  It didn't work in October, which is certainly troubling, but I'm still willing to give it the benefit of the doubt again, given the other positives that have been gathering over the past couple of weeks.

For the short-term, I mentioned in the Twitter notes and comment section of yesterday's note that the S&P was working on three straight "inside days" for only the second time since 1982.  When the market breaks down out of a triangle pattern like that, it has tended to rebound over the next few days to two weeks about 70% of the time.

This is something we've looked at several times over the years - the first move out of a volatility contraction has a pretty consistent tendency to be a "false" move.  The last occurrence in early October was not false - we just kept dropping - but the day prior to the inside days was a big down day, while this time it was a big up day.

We're also nearly through the worst part of December seasonality, and it starts to turn positive over the next few sessions through the first few days of the new year.  It failed last year, but historically the broader market has rallied 75%-80% of the time between the end of the second week of December through the first few days of January.  I don't rely too much on seasonality, but it's a consistent enough tendency to mention.

The support zones are clear here.  I want to see the S&P recover from above 850ish and the NDX from above 1140-1150ish, both of which we are currently testing pre-market.  The last-ditch levels would be 815 and 1100, respectively, but from the studies we've gone over during the past week, I would be surprised to see us drop that far and it would question the validity of the studies.  I'm going to bump up our intermediate-term signal strength to 50% bullish, based on what we've discussed over the past week.