Snapshot Notes - September 11, 2007

Yesterday morning, we went over the fact that our short-term models were on the cusp of giving buy signals, an event that should lead to positive performance over the next few sessions.

We got more selling pressure than I anticipated going into the yesterday's noon hour, but the indices recovered well and that continued today. So much so that those same indicators that were screaming "oversold!" yesterday were doing the opposite by this afternoon.

Given the relatively stretched nature of those guides, the great uncertainty regarding next week's FOMC decision, the looming overhead resistance from 1480 - 1500 on the S&P 500, and the potentially negative seasonality between Rosh Hashanah and Yom Kippur, I don't expect to see much more sustained upside from here.

I went over the past decade of seasonality readings related to those fall Jewish holidays, and recently the feared weakness between those two dates hasn't been all that terrifying. Still, it's something that other traders pay attention to and we should as well, at least from a psychology standpoint. If others are thinking that we'll see weakness here because of seasonality, then they just might be a little more inclined to sell, and with the other unknowns looming, I suspect there aren't going to be a rush of others looking to take the other side of their trades.

It seems to me as though the best bet from here is to not expect too much directional movement either way until next Tuesday is out of the way, and I'm expecting to see the S&P remain stuck between 1440 and 1480/1500 until then. I'd be even more inclined to sell should we see those STEM.MR Models cycle into overbought as the S&P approaches the upper end of that range, which may happen as soon as tomorrow if we get another decent up day.

As it does every year around this time, we hear the well-worn seasonal refrain "sell Rosh Hashanah, buy Yom Kippur". The nine days between the Jewish New Year and Day of Atonement are suppose to be weak, while the period following Yom Kippur is supposed to be the best of the year.

There is a little bit to this, but it's a stretch to assign blame or credit to the Jewish holidays. Volume does tend to thin out a bit around the holidays themselves, however that does not mean the directional bias is due to that either.

The table below shows how the S&P 500 performed over the past decade in the days between the holidays, then from Yom Kippur through the end of the year.

YEAR	BUY R.H. SELL Y.K.	BUY Y.K. SELL YEAR END
1996	1.1%	6.7%
1997	0.7%	0.4%
1998	-0.7%	20.9%
1999	-1.2%	10.0%
2000	-2.4%	-5.8%
2001	-1.4%	12.7%
2002	1.4%	-1.3%
2003	3.8%	7.5%
2004	-0.9%	9.2%
2005	-4.0%	6.0%
2006	1.6%	6.2%
Avg	-0.2%	+6.6%
% Pos	45%	82%

We can see that on 6 of the 11 years, the S&P was down in the period between the holidays, while on 9 of the 11, it closed the last part of the year in positive territory.

This is a bit spurious, simply because we're comparing a 9-day stretch to about a 50-day one. And with the secular rise in stock prices over longer horizons, it's no surprise that a couple of weeks in mid- to late September will be weaker than the most bullish quarter of the year.

I don't think there's anything particularly frightening about the upcoming Rosh Hashanah holiday as far as seasonality goes - we all know September tends to be the weakest month of the year and this is just a reflection of that.

In the meantime, the indices have managed to push higher throughout the day, and that has fully relieved the short-term oversold conditions we hit yesterday. In fact, we're starting to see some of the opposite extremes already, with all of our short-term guides in overbought territory except for the Cumulative TICKs and the STEM.MR Models.

I had been thinking that we would likely work higher in the short-term based off yesterday's buy signals in those models, but I've also been leery about expecting too much of a directional move ahead of the upcoming Fed decision. That's still a week away, and it's hard for me to imagine that many traders are going to want to commit capital ahead of that unknown, especially during a questionable seasonal stretch. So I'll be paring back my upside expectations from here, looking for more chop than sustained upside, especially if the S&P 500 cash index manages to make it up to 1480ish.

Good Tuesday morning...we begin the day with a gap up open in the major indices on mostly positive foreign markets. Federal Reserve Chairman Bernanke will be giving a speech in Germany this morning, but the chances he says anything market-moving seem relatively slim.

Yesterday we got a little follow-through from Friday's weakness, and a subscriber asked whether it's true that weakness on a Monday, following stronger weakness on a Friday, portends even more selling pressure going forward.

Checking the results, I was somewhat surprised to find quite the opposite. In the 28 instances in the history of the S&P 500 tracking fund, SPY, buying Monday's close in such cases and holding for the remained of the week resulted in 21 winning trades (75%) with an average return of +1.3%. The last nine instances, going back to October 2002, were all positive, as were 13 of the last 14, dating to April 2002.

Taking a little different tack, if we also stipulate that long-term bond yields declined by 3% or more over the following week as well, then we are reduced to only five occurrences (even going back to 1965), however I thought it was interesting that all five led to a positive market over the next week, and the average return was a large +3.5% with an average drawdown of -1.6% compared to an average maximum gains during the week of +5.0%. Without one particularly nasty week (right before the low in late July 2002), the average drawdown would have been only -0.6% in the four other cases.

Sharply declining bond yields have frequently been a bullish backdrop for stocks, and not just on a short-term time frame. For example, any time the yield on 10-year Treasury Notes have dropped by 10% or more over a two-month period (as they have recently), then the future quarterly return for the S&P 500 averaged +5.8% with 77% of days showing a positive return, and an average maximum gain more than twice the average drawdown. That's significantly higher than random three-month returns of +1.9% and 63% chance of being positive.

Back to the near-term, we got buy signals from both STEM.MR Models yesterday as they curled up from oversold territory, and the S&P 500 cash index bounced right off the 1440 area that looks like it's going to define the lower boundary of the current trading range.

We all know that the #1 focus of traders and investors alike at this point is the upcoming FOMC decision, and I don't know if we're going to get any kind of meaningful directional move ahead of that uncertainty. Even so, I'm going to fall back on the idea that we're going to recover at least a bit more from yesterday's oversold conditions and should see more of a positive drift in the coming days.

Forwarding or otherwise distributing this copyrighted material is a breach of your subscriber agreement. Violators are subject to termination of their subscription with any received subscription fees forfeited. Any references to historical performance are based on data we deem to be reliable, but are based upon feeds from third parties. We do not recommend subscribers take positions based on data presented here alone, but rather incorporate it into a comprehensive investment outlook.