Snapshot Notes - September 24, 2007

As of:

SPX 1519

During the past several days, I've been harping on the idea of negative seasonality following option expirations. We took a look at the data numerous different ways, and they all pointed to likely weakness, primarily concentrated very early in the week.

That seasonal bias made me hold off on what otherwise looked like a good long-side short-term setup Thursday afternoon. It's unusual for me to give seasonality priority over other technical and sentiment setups, but this one was consistent enough to give pause.

Now that negative post-expiration bias is mostly expired itself. There has been no consistent evidence of the weakness spreading to the rest of the week, unless we only look at recent September expirations. In that case, the last four years have shown some additional selling pressure for the remainder of the week.

The interesting thing with that, though, is that the next week showed very positive returns each of those years as well, usually more than erasing whatever selling pressure was seen following expiration. So even though there was some additional weakness for several days following expiration, that weakness was only temporary.

Earlier today, I went over the intraday reversal in the Nasdaq 100 while the S&P 500 was negatively diverging by being at least 2% below its own yearly high. I went into the study with the expectation of seeing further weakness at least in the short-term. As often happens, though, I was surprised when I actually looked at the hard numbers.

As opposed to preceding further weakness, these kinds of reversals led to positive returns in the NDX over the next week more than 70% of the time. More impressively, the following month showed a positive return nearly 90% of the time and the average return was a commendable +6.8%.

While I would not necessarily use that kind of stat as an excuse to buy, it does make me leery of getting overly negative here. Reversals from a high look ugly on a chart, but when we check history, that hasn't proved out.

The broader indices have been stuck in a tight wedge the past few sessions, so we could see a quick follow-through spike when prices trade outside that zone (as resting stop orders get triggered). I've been thinking that a move towards 1500 in the S&P 500 cash index should set up a high-probability long-side trade, so if we happen to see some additional selling pressure towards that level if this little wedge breaks to the downside, then I plan on more aggressively pushing on short-term longs.

From an intermediate-term standpoint, I still don't see much that would suggest we're going to roll over into a major decline from here. Most of our longer-term guides have cycled back into neutral from severe oversold conditions, taking away that fuel, but we're not seeing the kinds of excessive optimism readings that normally precede trouble. In fact, as I went over in a new Chart in Focus VIDEO today, there is more cash on the sidelines now - when compared with debt - than any month since 2004.

As of:

SPX 1519

HELP ARCHIVE

For the past couple of days, the theme I've been concentrated on involved weakness following an option expiration. Looked at in a number of different ways, we saw a consistent pattern of selling following expiration Fridays.

Now that we're getting Monday out of the way, that seasonal bias pretty much washes away. Anytime expiration Monday did what it was "supposed" to do and showed a negative return, the rest of the week was positive 61% of the time since 2002.

It hasn't been quite as rosy for September expirations specifically. Those showed a negative return for the rest of the week (the Tuesday through Friday following expiration) for the past four years. In 2003 the S&P's return -2.5%, 2004 was -0.9%, 2005 was -1.3% and 2006 was -0.5%.

Interestingly, though, buying at the end of the week and then holding for a week resulted in a return of +3.9% in 2003, +2.1% in 2004, +1.0% in 2005 and +1.2% in 2006. Those pretty much wiped away the losses from the post-expiration week.

Because of the afternoon selling, we've seen quite a reversal in the Nasdaq 100 after it squirted to a new high this morning. I checked for any time in the past 20 years when the NDX hit a new 52-week high intraday, then closed below its prior high, while the S&P 500 was more than 2% below its own high (which highlights the current divergence between the two indexes).

Surprisingly enough, the NDX showed positive returns going forward. A week later, it was positive 17 out of 24 times by an average of +1.9%, and a month later it was higher 21 times (88%) by an average of a whopping +6.8%.

Frankly, that shocked me - I would have thought that a reversal from a new high with a negatively diverging S&P would have showed solidly bearish results going forward, but it was quite the opposite.

I don't have much here that would suggest pushing any short-side bets is a high-probability trade, so my interest still rests with find a good long-side entry. Our short-term guides have backed off again with today's weakness but so far the only real oversold reading we're getting is from the Price Oscillator.

I'd still like to see a move towards 1500 to reduce the risk on a long trade, especially with the September expiration stats I noted above, but I don't want to keep being put off by seasonality stats that are now becoming a bit more dubious in their consistency when looking beyond the past few years. I'm going to move our position to 25% long at this point and will look to bump that up if we get more of a decline in the next couple of days.

As of:

SPX 1529

HELP ARCHIVE

Good Monday morning...we begin the next week with a slightly positive open indicated by the futures markets. Gap up opens following an option expiration tend to lead to pretty consistent behavior for the rest of the day, which we discuss a bit later in this note.

After the FOMC rate cut last week, I was looking for a good spot for a short-term long-side trade. Given what was still a positive intermediate-term outlook that was confirmed by some of the stats we saw for Tuesday's trading, the first short-term retracement should have offered a good setup on the long side.

I was looking for oversold conditions from our short-term guides and/or a move in the S&P 500 cash index back towards 1500ish, but I was not interested in immediately chasing prices higher after Tuesday's jump - markets have a tendency to settle down for a few days after moves like that.

We got enough of a pullback on Thursday to trigger an oversold signal from the STEM.MR Model for the S&P, which looked like it should have been a very good spot to try that short-term long-side trade. I hesitated for one reason, however - seasonality.

On Thursday afternoon and Friday, I went over a bunch of different looks at the S&P's performance around option expiration days. All of those studies pointed to a risk/reward ratio that was skewed greatly to the short side. So we had a battle between sentiment (which was suggesting higher prices) and seasonality (which was suggesting lower ones).

I very, very rarely let seasonality trump anything else I watch, since it tends to not be a consistent enough edge. But those stats surrounding expirations were just too overwhelming for me to ignore. In fact, here's another one - buying the open on any gap up Monday morning following an option expiration and holding until the close has resulted in only 28% winning trades (9 out of 32) since the bear-market low in October 2002, with an average return of -0.2%. That includes four losers out of five attempts in 2007.

On average, the losers didn't gain more than +0.19% from the open before reversing to close lower. Of the 23 losing trades, 18 of them didn't gain any more than +0.25% from the open during the day.

For the days that bucked the stats and closed higher, they averaged an intraday drawdown (i.e. loss) of only -0.16%, and only two of them lost any more than -0.20% during the day.

For today, assuming the indices open about where they are now with the S&P futures up 4 points, that means that the index shouldn't gain any more than about another 4 points from the open, which would take the cash index to around 1533 - which also happens to be Friday's high. A breakout above that would not only be a "breakout", but it would also violate almost all the negative precedents given by these option expiration stats. If the index loses more than 4 points or so from the open, however, it would help to validate the negative connotations from the opex studies.

So those are the general thresholds I'll be watching for this morning to see if it's likely we're going to get that weakness today that seasonality predicted last week. If not, and we get a sustained breakout over last week's highs, then I'm going to be an unhappy camper for letting seasonality talk me out of a good setup from last week. But I don't chase prices, so I would have to sit back, stew over a missed trade, and wait for the next high-probability opportunity.

Most expiration-related weakness tends to be heavily concentrated early in the week, particularly Monday. So after today, I promise I'll get off the "negative expiration seasonality" thing and return to a more normal course of business.

On a quick side note, if you've emailed us about the lock-in offer and have not yet received a response, please bear with us a bit...you should receive one in the next day or two. We will of course honor any request we receive before Oct 1, even if you don't get an immediate response.

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.