Snapshot Notes - November 7, 2007

As of:

SPX 1511

The basic philosophy with which I approach trading is to buy oversold conditions and sell overbought ones. The knock against that kind of approach is that oversold can become more oversold, and overbought more overbought.

That's very true, and as with any strategy there are going to be losses. The key is in trying to find ways to minimize them (both in frequency and magnitude), while still preserving the opportunity to profit.

For overbought/oversold types of trading, good ways to minimize those losses are to be selective and rigorous about what is considered "extreme", and to consider the overall market environment as context. So basically, I look for different types of confirming evidence that an extreme has been reached, and become more aggressive when that extreme occurs in line with the current trend (i.e. more aggressively long when we're oversold in an up-trending market, and more aggressively short when we're overbought during a down-trending one).

That helps to generate enough activity to make it worthwhile, while usually keeping losses small and relatively infrequent. Once in a while, we see enough evidence that it makes sense to take a flier when we reach an extreme, period, regardless of the overall trend. One of those looks like it's approaching in the Banking sector.

Earlier today I went over the fact that the S&P Banking Sector Index had dropped more than 4% today, and more than 20% from its recent 52-week high. That's the worst one-two punch in five years, and historically has come before a trade-able low in the sector.

When we've seen this kind of selling pressure during the past 18 years, the three-month return in that index was positive 17 out of 17 times, by an average of +15%. The average drawdown (i.e. maximum loss during the trade) was not insignificant, at -7%, but it was still dwarfed by the average maximum gain during those three months of +20%. On average, it took 23 trading days for the index to find a bottom, suggesting that sometime this month we should see that sector find a trade-able low. I'll be watching for an intraday or consecutive-day reversal to signal that perhaps we've reached an exhaustion point.

On a shorter-term basis, I've been struggling to find any kind of extremes among our more sensitive guides or studies, and have been mostly frustrated in those attempts. Despite the selling pressure of the past week, we haven't seen many "excessive pessimism" types of suggestions popping up, which has kept me on the sidelines. I've been most interested in looking for oversold conditions, based on the idea that we're in the midst of a seasonally positive time of the year, and the technical picture in the major indices was still quite positive.

That's changed a bit with today's trading. With a close under 1490, the daily chart of the S&P 500 cash index now clearly shows a progression of lower highs and lower lows over the past month, and the 50-day moving average is in danger of rolling over. Those basic tools are what I look for to determine the intermediate-term trend, and if they turn negative then I'll become much more interested in selling into overbought conditions, particularly given the concerns I went over in a comment this past weekend, and also (less importantly) the cluster of "Hindenburg Omens" that I wrote about this morning. But we need to see some extremes first in order to define some kind of edge, and so far I'm still not finding many.

As of:

SPX 1511

HELP ARCHIVE

A couple of weeks ago, I mentioned that as far as I was concerned, the banking sector was in a bear market, with the BKX index under declining 50-day and 200-day moving averages. Short-term bounces will happen off of extreme moves, but overall the momentum was clearly to the downside.

At the time, I mentioned something that suggested that a bounce in the index was likely (it had been down for seven consecutive days), which it managed to do for a few days. Today we're getting something that is equally as interesting, and perhaps for more than just an ultra-short trade.

The S&P Banking Sector Index, for which I have more historical data and which is very highly correlated to the BKX index, will have shed 4% of its value today alone, and more than 20% from its 52-week high, should it close where it's now trading. I have data on this index back to late 1989, and checking for any time we've seen this kind of combination, we were often within weeks of a notable low.

The three-month return in the index averaged +15.0% going forward, with 17 out of 17 occurrences displaying a positive return. It wasn't easy, though...the average drawdown (i.e. maximum loss) during that time was -7.0%, but at least the eventual reward made up for it - the maximum gain during the next three months averaged +20.4%. On average, it took the index 23 calendar days (about three weeks in trading days) to form at least a one-month low in which the index subsequently rallied at least 5%.

So based on about 20 years of history, the banking sector has been hit hard enough that it's beginning to look interesting. Short-term risk is high, there's no doubt about that - while the index bottomed twice on days like this, the majority of time there was more pain to come over the next two-three weeks before a tradeable low was formed. Once the low was put in, though, the upside was notable indeed.

If it follows history, then we should see a good low in the sector this month. According to the Rydex and ProShares asset flows that we track on the site, traders have given up on the sector to a degree that would support a good low once interest in the group began to pick up again. I'll be watching for a major intraday or consecutive-day reversal in the BKX or other banking-related indices to give the sign that perhaps we've hit the washout point sometime during the next couple of weeks.

And a low in the financial indices would be good news for the broader indices, of course, especially heading into the year's rear. We have seen sporadic instances of short-term oversold readings in indices like the BKX and XBD, which for the most part have played out over a very short-term time frame, but now we're getting to a point that's more interesting from a longer-term point of view.

I mentioned a stat this morning that suggested we not try to buy into today's big gap down open, particularly if the S&P later went on to hit a new intraday low after the first hour. Indeed it did, and the index got smacked in response, losing more than 2% on the day before the recovery attempt of the last hour.

Our intraday guides are still mostly neutral other than the Cumulative TICK for both the S&P 500 and Nasdaq 100, and the S&P has been able to regain the 1490ish area that so many are watching, so there is some potential for a stand here. I don't particularly like the looks of things from a technical basis (we're staring at the potential for lower highs and lower lows on a daily chart of the S&P), and from a sentiment perspective it's hard to find many suggestions that we have excessively pessimistic conditions. I continue to believe that we're in a choppy market where buying oversold and selling overbought conditions is the way to go, but I'm still struggling to find either.

As of:

SPX 1511

HELP ARCHIVE

Good Wednesday morning...we begin the day with a slap upside the head in the pre-market futures, with the major indices down around 1%. You can pick the reason behind the early move, be it rumors of yet more trouble in the financials, or the looks-like-a-waterfall drop in the US Dollar.

This is the second very weak open in two days. This is the kind of gap risk I don't like to take, though usually it's more common during the heat of earnings season. The last time we saw this kind of opening gap volatility (such as we've seen during the past three days) was October 2002, so it's been awhile.

According to Jason Roney on Minyanville.com*, the S&P 500 futures have shown this pattern five other times, and in five of the six cases, the futures ended up losing more than 2% on the session. My tests showed that several of these occurred during September/October 2002 when volatility was at its peak.

One of the "market crash" signals that has become a fairly popular topic of late is the Hindenburg Omen. I've written about this indicator a few times over the years, including a couple of times in the past month, and it has a relatively successful record at preceding short- to intermediate-term weakness in the general equity market - but it's far from a reason to expect an all-out crash.

The indicator tries to identify times when the broad stock-market indices are doing well, but there is a wide split among individual stocks, with many of them hitting new highs on the same day that many are hitting new lows. It's a rare situation, though it has been triggering consistently over the past month. Different analysts will come up with different figures, but I show that there have been seven confirmed Hindenburg Omens during the past 30 days.

The last time we saw this concentrated of a cluster of Omens was late April 2006, which was an excellent early-warning sign of looming weakness in equities. Overall, I can find five such clusters over the past 40 years, and all of them were successful to varying degrees at highlighting times of increased risk. After all five instances, the S&P 500 was lower 30 days later. For those curious, the dates I have are 06/01/65, 06/10/71, 03/30/72, 12/13/99 and 04/26/06.

As for the shorter-term, we have become accustomed to buying into large gap downs and selling into large gaps up. There's good reason for that - stocks usually move against extreme opening moves. In various contexts, though, that doesn't work as consistently and per Jason Roney's comments above, I wonder if perhaps we're about to see that today.

As Jason notes, trading during the first hour is often an excellent guide to the rest of the session - if the S&P makes a lower intraday low after the first hour, then we might as well write off the idea of a big intraday rebound...it just doesn't often happen. But if we can bounce from the opening gap and hold the initial low, then we have a much better chance at seeing higher prices. The 1490-1500 support zone in the S&P cash index is also coming into play here, so obviously the bulls wouldn't want to see that go by the wayside.

I don't have a solid reason to look for either reaction - as I've been noting during the past week, it's been tough for me to find a reliable edge in this market, and this kind of chop is difficult to navigate except on the shortest of time frames. I'm still staying mostly out until I can identify something that seems to have a higher probability of success.

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.