Print Comments

TUESDAY, MAY 6, 2008

Another Volume Consideration

05/06/08 1:50 PM EST

As of: SPX 1375 HELP  ARCHIVE

After some early-morning weakness, stocks have turned and are trying to carve out a little bit of a trend day, where stocks open at their lows and close at or near their highs of the day.

We haven't met the classic definitions of a trend day that have worked so well on a handful of occasions so far this year, but still we're seeing consistent buying pressure when the NYSE TICK approaches the zero line, and we're hitting new intraday highs in the indices when that happens.  We're seeing that pattern crack a bit heading into the late afternoon, so I'm not counting on a close near the highs as much as I do when the indices more cleanly meet the conditions for a trend day.

Given the news flow and other metrics we've discussed, I'm more than a little bit surprised by this behavior.  While we had a smattering of minor oversold indicators coming into today and the negatives we'd discussed were minor, I didn't think this morning's gap would turn into a trend day (at least to the upside).

Over the past couple of weeks, I've mentioned volume more than I probably have in the past couple of years combined.  Other than when we see very high turnover, I typically ignore it.

But we're seeing some exceptionally rare developments with regards to volume, in terms of how low it has been on an absolute and relative basis.  Outside of holiday periods, we very rarely see turnover this low.

I want to touch on one other aspect of volume, this time in terms of the Nasdaq relative to the NYSE.  One of the charts we update on the site is a Nasdaq/NYSE Volume Ratio, which compares volume between the two exchanges.

Basically, when the ratio is high, it shows a lot of speculative preference for issues on the Nasdaq; when it is low, it shows a flight to the "safe" stocks on the NYSE.  From a contrary point of view, the former should be a negative for the markets, the latter a positive.

To check that out, I looked at the data we have going back to 1996.  I looked for either extreme, and calculated how the S&P 500 fared over different time periods going forward.  First, let's look at the times when the ratio was 1.85 or higher, which is what we're seeing now.

There were 71 days that met this condition, and going forward it didn't bode all that well for the S&P.  By three months later, less than a quarter of the days sported a positive return, and overall the days averaged more than a 5% loss over the next few months.

Now let's look at the opposite situation, when volume on the Nasdaq was low compared to the NYSE.

There were 113 days that met this condition, and the results going forward were exceptional.  Three months later, more than 90% of them showed a positive return, which averaged a massive +9.7%.

Clearly, the outlook for equities has been better when volume has shifted to the NYSE as opposed to the Nasdaq.  That makes the current situation concerning, as the ratio as of yesterday was 1.89.

I do have one caveat about this data - it will greatly vary depending on who your quote vendor is.  I checked my data against the NYSE, the Wall Street Journal and Bloomberg, and came up with different values each time.  They all showed that Nasdaq volume was high relative to the NYSE, but not necessarily to the degree of the Reuters data we use for the chart.

We do use the same data now as we've used historically, though, and as far as I know Reuters has not changed their volume calculations, so the current reading should be comparable to the historical data referenced in the tables above.

So once again we have a potential negative from the current volume situation, which comes on the heels of the other minor negatives we've already discussed.  Given our overall downtrend, lack of volume, negative breadth divergences, upcoming negative seasonality as we went over this morning, and spotty signs of too much enthusiasm from traders, I'm becoming increasingly cautious on a short- and intermediate-term basis.

In looking for a solid setup - an edge that suggests it makes sense to actually risk capital to capitalize on those thoughts - I'm still coming up mostly empty.  We're just not seeing price weakening much, and the price-based and other patterns I've looked at have not suggested a strong downside edge.  So I'm still mostly waiting, and continually looking for that edge that will suggest it makes sense to bet against the rally.

Seasonality Index is a Cause for Pause

05/06/08 10:00 AM EST

As of: SPX 1375 HELP  ARCHIVE

Good Tuesday morning...We begin the day with some selling pressure in the major indices and most of the broader sectors.  We're seeing a sector pattern that was prevalent during most of the January - March decline, with the "evil twins" of Gold and Oil leading to the upside, while Financials, Retail and Housing get sold hard.  If this continues through the day, there's little chance we'll see an upside reversal in the major indices.

One of the features we post on the site is a Seasonality Index created by Jay Kaeppel.  The Index looks at several seasonal factors and assigns each day a score whenever one of them is present.  If none are applicable, then that day gets a score of "0"; if all of them are present, then that day gets a "4".  The higher the score, the better the chance for the market to rise.

This is becoming increasingly interesting to me since we're about to enter a long string of "0" days as you can see from the calendar on the site.  This will be the first time since the summer of 2006 - a tough time for the market - that we had a long string of "0" days.

We can check to see how this would have worked during the history of the S&P 500 tracking fund (SPY).  During its lifetime, SPY has gained just under 95 points.  Let's break that down by how many points it accumulated during each of the differently-scored days (the average return per day is in parentheses):

Seasonality Index = 0:  -29 (-0.1%)

Seasonality Index = 1:  +5 (+0.0%)

Seasonality Index = 2:  +73 (+0.1%)

Seasonality Index = 3:  +46 (+0.2%)

Seasonality Index = 4:  -2 (-0.1%)

From the scorecard above, we can see that the S&P performed pretty much in line with the scores - the lower the score, the lower the average return.  The one deviation is "4" days, supposedly the most bullish of all, which actually proved to have a negative return.  The reason is that there were very few of them (11 in total), and one day skewed the results.  8 of the 11 showed a positive return, but one bad boy of -1.6% in 1999 ruined the overall positive expectancy.

If you would have been invested just on days scored "2" or "3", you would have gained about 120 points in SPY with a maximum drawdown of 25 points.  That compares to a buy-and-hold gain of 95 points with a 75-point max drawdown.  Clearly, paying attention to the Seasonality Index would have made a huge difference.

Of course, that means a lot of trading, which would drastically reduce the results if we include slippage and commissions.  I'm not trying to put forth a trading system here, rather I want to see whether it would have paid to be more bullish on some days than others, and that seems to be the case.

For the Nasdaq 100 trust (QQQQ), the results were equally dramatic.  Here's the breakdown by score:

Seasonality Index = 0:  -30 (-0.1%)

Seasonality Index = 1:  -26 (-0.0%)

Seasonality Index = 2:  +35 (+0.1%)

Seasonality Index = 3:  +15 (+0.3%)

Seasonality Index = 4:  -1 (-0.2%)

If we would have been invested only on "2" or "3" days, then we would have gained a total of 52 points in QQQQ, with a maximum drawdown of only 18 points.  If we would have bought and held QQQQ since inception, then we would have shown a loss of 3 points with a maximum drawdown of a whopping 98 points.

Again, I'm not trying to push a trading system.  What I want to do is figure out whether the upcoming string of "0" days is cause for concern, and I believe it is based on the data above.

That means that I will be looking to take more aggressive short-side positions when the setups exits, and lighten my exposure to the long side in a general sense.  Seasonality is a tertiary indicator to me, but it does influence my trading, especially in terms of trade sizing.

Over the past week or so, we've gone over some minor negatives.  Those included the recent bout of exceptionally low volume, a burst of optimism from Rydex traders and individual investors, and most recently a big breadth divergence.  All of those factors point to neutral to lower prices going forward, but I would consider them all to be minor points only.

They are not the type of data that gets me excited enough to want to push aggressively on the short-side.  The S&P 500 had been doing quite well holding up above that 1400ish area that had proved to be resistance until the recent breakout.  My feeling was that it wouldn't hold, and this morning it dropped back below, but it's way too early to suggest that we've seen a false breakout.

I continue to see mostly a lack of solid edges either way, but I'm still tilted to the neutral/slightly bearish camp.  We are seeing some oversold indications among our most sensitive indicators, which helps to reinforce the idea that this is a muddled picture here.  I think the potential negatives outweigh the positives, so I'm still looking for choppy to downward price pressure, but again I don't see anything that makes me want to press hard on any positions.

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.

© 2008 Sundial Capital Research, Inc.  All Rights Reserved.  www.sentimenTrader.com