Snapshot Notes - May 13, 2008


	Print Comments

TUESDAY, MAY 13, 2008

Just a Short-Covering Rally?

05/13/08 2:45 PM EST

As of:

SPX 1375

HELP ARCHIVE

I'm not a fan of trying to assign cause and effect to the market's movements, as there are usually too many moving parts to know with any certainty why the market's doing what it's doing. The few exceptions are when there is a clear, immediate spike up or down based off of a news event.

One of the more-popular reasons I've seen for yesterday's exceptionally low volume was that it was simply short-covering. That happens when traders sell short stock hoping for a decline, then buy the shares back to cover the short position. The assumption would be that if that was the case, then it's a weak foundation for more gains.

I'm not sure how anyone would know that yesterday's rally was due to short-covering, or why it matters, but I've received a few questions on it and thought I'd take a look.

If the rally was due to short-covering, then it would seem most likely that the index that put in one of the best performances yesterday, the Nasdaq 100, would have seen its most highly-shorted component stocks do the best, while the least-shorted should do worse.

To check, I looked at the current short interest ratio for each stock in the NDX, then compared that to how the stock performed yesterday and placed each stock into one of 10 bins based on how heavily shorted it is (see chart below) .

If the "short covering theory" was valid, then the bars on the left-hand side of the chart should be much higher than the ones on the right-hand side.

There's not much evidence of that in the chart. And it doesn't change if we use another measure of short intensity - the level of short interest as a percentage of the stocks' floats. Using that measure, the chart looks almost exactly like the one above.

Let's take one more look at the data, this time a scatter plot that more clearly shows the correlation between the short interest ratio for each stock and its performance from yesterday. If there was a positive correlation between the two, then the solid black line on the chart should slope up and to the right, with a tight grouping of little blue diamonds around it.

No such luck here, either. In fact, the correlation was slightly negative - the more-shorted stocks actually did a little bit worse than the less-shorted ones.

Does that mean that yesterday's rally was due to "real" buying and not just folks covering short positions? There's no good way to determine that with the information I have, and quite frankly I'm not sure if it even matters. The fact is that volume was low on an historical scale as we discussed this morning, and that usually isn't a good sign for the market no matter who was doing the buying (or lack of it).

Over the past week, I've noted a couple of times that inter-market relationships that we've been able to count on for the past seven months have started to get a little shakier. We're seeing that again today, as Treasury Notes are getting hammered, driving yields higher, but stocks are not benefiting.

Since the October high, there have been 11 days when the yield on 10-year Notes jumped by more than 3%, as they're doing today. On those 11 days, the S&P 500 closed higher 9 times by an average of +1.6%.

There were only 4 days that didn't show a gain in the S&P of more than +1% (12/7/07, 2/7/08, 2/14/08 and 3/5/08). There wasn't anything especially consistent about those days going forward - two led to almost immediate and stiff selling pressure during the next week, while the other two led to choppy upside.

I think the main takeaway from this is that we need to be aware that some of the relationships we've seen since October are starting to break down with increasing frequency. Perhaps this means that we're heading towards a more "normal" environment where stocks trade more on their own merits than on global macro forces.

Even so, we've discussed a few negatives over the past couple of days, related to the low volatility and even lower volume. Those aren't exactly encouraging developments, and I remain cautious here.

It's been an extremely frustrating couple of weeks lately, as I haven't been able to find a solid enough edge to want to risk much capital. Most people in this business are hard-wired to want to "do something", even if it's detrimental, in order to prove that we're doing something. It's a bad habit, though, which is what I keep telling myself during this mostly inactive period.

About the only potential trade on my radar at the moment is a short sale in the S&P if we see more of a meek rally attempt that doesn't take us over last week's high, or the 2000 level on the NDX and 740 on the Russell 2000. The timing on a short is questionable, as I can't find much that suggests we should see weakness right now. Wednesdays of option expiration weeks have been consistently positive over the past couple of years, but that's a weak argument for or against a trade, so I'm still working and trying to find something that could provide us with a good edge.

Lack of Concern in Volume, Volatility

05/13/08 9:15 AM EST

As of:

SPX 1375

HELP ARCHIVE

Good Tuesday morning...We began the morning with an indicated gap down in the major indices, but after a better-than-expected retail sales report and more confidence-building from Fed Chairman Bernanke, stocks have turned higher, bonds are down, oil is down, the Dollar is up and it looks like full steam ahead.

One aspect that's getting increasing attention is volume, or rather the lack of it. We've been looking at this for the past couple of weeks, which is unusual - I don't typically place much weight at all on volume. The exception is when we're seeing something unusual, and the current case absolutely qualifies.

Using Reuters data (and confirmed by Bloomberg), yesterday was the lightest-volume day of the year on the NYSE and very close to it on the Nasdaq. That's a remarkable statistic, especially considering the several occasions during the past couple of weeks when we've already highlighted low-volume conditions.

How unusual is this? Well, going back to 1980 with this data, it has never happened before. We have never seen the lightest volume in six months during May. It has really only occurred during holiday trading sessions. Going back 28 years, the "lowest volume in six months" has happened 65 times, with the following monthly distribution:

January: 0

February: 1

March: 0

April: 0

May: 0

June: 1

July: 12

August: 14

September: 3

October: 5

November: 19

December: 10

As you can see, the first six months of the year have shown only two days out of the entire sample when volume was the lowest out of the past six months. The February occurrence was in 1991, after which the market went precisely nowhere for the next nine months. The June instance was in 2001, after which we rolled over immediately into the teeth of the last bear market.

Looking for any time when the S&P has rallied 1% or more on the lightest volume in six months, there were 8 occurrences. Since most of them were around a holiday, trading was probably more influenced by that seasonality than anything, but over the next couple of days the S&P was up only 2 times and sported an average return of -0.7%, with a risk (i.e. maximum drawdown) more than twice as large as the reward (i.e. maximum gain). After that, the returns were mixed.

The one-month average of NYSE volume has now sunk 19% below the average of the past year, the greatest deviation since October 23rd of last year (just as the market was topping out). This is one of the largest deviations in nearly 30 years, with almost every other one even coming close occurring in August or September.

I have noticed an increasing number of oddities with volume lately. Over the past year, there has been a strong difference of opinion as to what "total volume" actually is. Sources like Reuters, Bloomberg and the Wall Street Journal have been carrying lower volume figures than sources like CSI or eSignal, which is why every time I write about low volume I get a few "no, it's not" emails.

Perhaps it is due to the counting of electronically-traded shares or some other structural reason, but the bottom line is that it shouldn't matter what vendor you use as long as you're consistent - and the vendor hasn't changed their volume calculations over time.

Yesterday morning, we discussed a few factors that were short-term bullish in nature, and it looked like we'd get some kind of rebound off the technical support levels that the major indices were approaching. The buyers really ran with it yesterday, and it ended up skewing some of the factors we look at back to being mildly bearish. One of our intraday indicators reached a reading that has led to trouble for stocks during the post-March rally, and we also saw implied volatility absolutely implode yesterday. Given that the S&P didn't even trade at a three-month high, but the VIX dropped to a six-month low, returns going forward were sub-par.

Heading into last week, there were a number of negative factors that we'd discussed, including low volume, speculation in the Nasdaq as opposed to the NYSE, negative breadth divergences, and a big drop in pessimism from individual investors. Last week's pullback alleviated some of those negatives, but yesterday's jump brought a couple more to the fore. I continue to think that we're going to have trouble making much sustained headway due to these factors, and will become more interested in betting against a continued rally as we rise. If the Nasdaq 100 and Russell 2000 can climb over and hold 2000 and 740, respectively, then I will back off that idea, but as long as they're not, and we have the conditions we're seeing now, I don't especially like our chances of seeing a big upside move that holds.

All the best,

Jason Goepfert

President and CEO

Sundial Capital Research, Inc.

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.