Snapshot Notes - September 28, 2007

For the past week, I've been something of a slave to seasonality, which is atypical. First it was the tendency surrounding option expiration, then quarter-end, which overrode the other types of analysis we're more prone to study.

Now we're heading into a more "normal" environment that won't be subject to such consistent seasonal biases. There has been a relatively positive start to October when September bucked its trend and closed well into positive territory, but otherwise there isn't much to note.

Aside from seasonality, the shorter-term guides that we follow are all firmly in neutral territory due to the chop of the past several sessions. Price-wise, the S&P 500 cash index has a slug of resistance around 1535-1540 which marked the June highs, the breakdown in July, and the post-Fed high from last week. There should still be strong support around 1500ish, so those are the main technical levels I'm watching here.

More interesting to me would be a pullback next week that takes us again towards 1500, if accompanied by some oversold readings from our more sensitive indicators. I think that'd present us with a better situation from the long side.

The latest Commitments of Traders report showed that large commercial hedgers (aka the "smart money") made a big change in their futures contract positions for the first time in months, helped along by the expiration of September contracts.

They went from a record net long position of $42 billion all the way down to $18 billion, due to a huge shift in S&P 500 e-mini and large Nasdaq 100 contracts. That $18 billion net long is still very large when looking at the past seven years or so, but it's a big change from the past couple of months. I've been doubting the validity of this data for many weeks now, and perhaps we were seeing something "goofy" with the September contracts. If so, then we should see more normal reporting going forward, which is hopefully the case.

I've spent a lot more time than usual on seasonality lately, either related to option expiration or the looming end to the quarter.

I'm not that big a seasonality fan, but the tendencies we've seen surrounding those events have been consistent enough to note. I've been asked quite a bit about what to expect now seasonality-wise, so I want to go over that set of stats, then leave the seasonality concept alone until the holidays when we start to see more consistent readings again.

September has the reputation for being nasty, and for the most part it has been deserved. As you can see from the Seasonality section of the site, the month has both the lowest average return (by far) and the lowest percentage of time positive of any month.

So what happens when the month bucks its reputation and closes well in the green? I checked for any time the S&P 500 rose more than 3% during the month of September, and in the following table put its returns for the next few days, weeks and months.

Look-Ahead	% Positive	Avg Return	Max Gain	Max Loss
Next 5 days	91%	+1.4%	+2.1%	-0.7%
Next 10 days	73%	+0.8%	+2.6%	-1.1%
Next Month	64%	+0.4%	+3.1%	-1.3%
Next Three Months	73%	+4.3%	+7.3%	-3.9%

The data was very positive in the short-term, with 10 of the 11 years showing a positive return through the first five trading sessions of October (and much of that gain came in the first few days). Only three of them showed a maximum drawdown during the week that was worse than -1%.

Most of the positive drift was out of the way by the first week, though, as the average return dropped the further out we looked during the month. The three-month return was quite positive, which is a function of the last quarter normally being the best of the bunch.

I also checked the same thing for the Nasdaq 100, but the results were less conclusive. There were five years in the past twenty-two that qualified, and the returns during various stretches of October were mixed and in line with random.

OK, enough with the seasonality stuff. It could/should be a mild positive for the broader indices like the S&P given that September was so positive, but other than that there's not a whole lot to ponder.

Back to our more normal course of analysis, the shorter-term guides that we follow are all firmly in neutral territory due to the chop of the past several sessions. Price-wise, many of the indices have coiled into tight intraday ranges the past couple of days, which tends to resolve to the downside, particularly when overbought at the time. That would most apply to the Nasdaq 100 at this point as opposed to some of the broader indexes.

The S&P 500 cash index has what looks to be probable resistance around 1535-1540 which marked the June highs, the breakdown in July, and the post-Fed high from last week. There should still be strong support around 1500ish, so those are the main technical levels I'm watching here.

I am not a breakout trader by habit - the risk/reward of those trades in broad indices makes me uncomfortable - but given the potential positive bias in the early part of October, if the S&P can make it over that 1540ish hump, then I'll look at taking a smallish long position to see if it can carry to new highs next week. The intermediate-term stuff we've looked at is still at least moderately positive, so there's still some breeze at our backs in that respect. Most interesting to me would be a pullback next week that takes us again towards 1500 - if accompanied by some oversold readings from our more sensitive indicators, I think that'd present us with a better situation from the long side.

Good Friday morning...we begin the end to the week with some slight weakness in the major indices and mostly negative performance in most of the broader sectors. Banking and Housing are taking up their now-familiar spots at the bottom of the totem pole, while Retail is enjoying what has proved to be a rare respite.

For the past several sessions, I've been concerned about the broader market's potential to sustain any further gains due to the constant game of sector rotation that we were witnessing. Technology had been one of the few standout sectors, and seemed to be carrying some broader indices like the S&P 500 along for the ride despite some notable weakness in Retail, Housing and Financials. Sustained moves in either direction in the S&P tend to happen when most of the "important' sectors are traveling in the same direction.

In addition, some of the strong tech indices like the Nasdaq 100 were displaying extreme overbought conditions, with narrowing intraday ranges. That kind of combination in the past has led to consistent weakness going forward, at least in the short-term, and made me less inclined to trust that sector's ability to continue to shoulder the broader market's burden.

On Wednesday, when it seemed like we were about to roll over, rumors of Warren Buffet buying a major stake in a broker propped us up into the close. Past instances of those kinds of rumors have led to a wicked hangover for short-term traders, and indeed Bear Stearns is off nearly 5% from the height of Wednesday's buying frenzy.

Without those kinds of shaky props, heading into what has traditionally been a relatively weak seasonal period (the last days of a quarter), I didn't see a good risk/reward situation on either side of the ball, and that's still the case. The indices have done nothing "wrong" at all so I don't see the sense in trying to short, but for an index like the S&P 500, the sectors that have propped it up seem unlikely to keep doing so for the time being, so longs aren't particularly attractive here either. That means I'm flat for trading positions and waiting for a better setup.

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.