Snapshot Notes - May 15, 2008


	Print Comments

THURSDAY, MAY 15, 2008

I'm Not a Believer, But The Persistence is Impressive

05/15/08 4:15 PM EST

As of:

SPX 1375

HELP ARCHIVE

Coming into today, many short-term traders were keying off yesterday's reversal bar in the major indices, assuming the technical pattern was a reason to assume that buyers were exhausted.

Like volume (in most cases), reversal bars are more hype than help. This morning I mentioned that I tested yesterday's move a dozen different ways, and all of them came up empty in terms of suggesting we'd see more than perhaps a few hours of follow-through to the downside. Which was too bad, because I'm positioned net short.

My reasoning for betting against a sustained rally has been that the indices had already met most of the projections from the studies we went over in March, we are still mired in a long-term downtrend (a downward-sloping 200-day moving average), we've been witnessing an historical drop-off in exchange-wide volume, implied volatility has fallen faster than justified by the move in the underlying indices, seasonality is the worst since the summer of 2006, we're seeing a smattering of "excessive optimism" readings, and the major indices were all below what should prove to be resistance.

That's quite a list, but we can now cross off one of the negatives, at least for most of the major indices. The recovery today took us back over the levels I (and probably most everyone else) was watching, which are 1420 in the S&P 500, 2000 in the Nasdaq 100 and 740 in the Russell 2000. While such obvious levels might conjure up the possibility of a "false" breakout to clean out buy stop orders resting just above those areas, I have a hard time having any conviction in such conjecture. As far as I'm concerned, the breakout is legitimate unless we see a rejection and trade back below them.

The move to new highs brought with them a few short-term overbought readings, such as the Price Oscillators moving well above 60% for both the S&P and NDX. Similarly stretched readings have coincided with the indices stalling out during the past couple of months, so the bears have a sliver of hope there.

As noted this morning, our Dumb Money Confidence is now at 71%, the worst reading since January 2007. This is a special kind of overbought, as it has been rarely achieved during a generally downtrending market. Even during the big bull run from 2003 - 2007, we rarely saw this level of Dumb Money conviction, and it had a pretty good record at highlighting times when the indices were moving into the flattish, choppy upside stretch of their moves, and which were ultimately given back during sharp, short corrections.

I also posted a Data Brief this morning that highlighted the level of speculation in the Nasdaq among Rydex traders. They've moved money into the leveraged long funds at Rydex to such a degree it rivals the most-speculative times of the past five years. As laid out in the Brief, forward returns after past instances in the Nasdaq were...uhh...subpar, to put it mildly.

All of this means that I have a very difficult time chasing prices higher here, as the probability of these gains being given back seems quite high. But being positioned net short as of yesterday morning has made me exceedingly uncomfortable given today's late breakout. No matter how strong the edge is that I think I might have, I will not forsake risk control. If we continue to hold above the aforementioned levels on the major indices, then as a matter of discipline I'll have to reduce exposure or excuse myself from the short side, no matter how wrong that may feel at the time, and look to re-enter if/when prices show some better signs of weakness taking hold.

"Sentiment Alpha"...Really!?

05/15/08 9:25 AM EST

As of:

SPX 1375

HELP ARCHIVE

Good Thursday morning...We begin the day with a very slight bump up in the pre-market futures. There was some merger activity overnight, which always gets futures traders excited, but otherwise we had lackluster economic reports and mixed news overseas.

Over the years, I've fielded a number of questions about whether it's possible that by maintaining this site, we're sealing our own doom. In other words, by promoting the use of sentiment as a discipline, are we shooting ourselves in the foot by impacting the indicators we follow?

I've brushed off those questions for seven years. The readership of this site has been concentrated enough that it hasn't been much of a worry to me, and over the years I've seen first-hand just how mis-trusted sentiment is as a discipline. It doesn't matter how often fundamental and technical analysis fails - they will always be a part of almost everyone's toolbox. But there is very little tolerance for sentiment failures.

Sentiment has also gotten no respect from the academic community. Frankly, I've welcomed that because they've typically approached it from the wrong way time and time again. But that is starting to change in a big way. We are seeing more and more studies showing how using sentiment as an input can increase returns over a benchmark, or "alpha" as it's called. Read this great post from the All About Alpha blog for the latest example.

A trend that we've seen accelerate recently is academics publishing a study on a certain discipline, then a horde of copy-cats moving into private money management based on those principles. With the increasing number of studies using sentiment as an efficient tool, we're probably going to see some changes. I touched on this effect in a Data Brief regarding short interest earlier this year.

This by no means suggests that all of a sudden the [insert favorite sentiment indicator here] is going to become worthless tomorrow. But it does mean that we may have to become a little less tolerant if some indicators begin to act "weird". When we see them behaving in a way that is counter to how they typically act during certain market condition (cough, cough...Commitments of Traders data), then we should cast a wary eye towards weighting that data too heavily in our analysis.

As for the here-and-now, we had a highly unusual reversal yesterday afternoon. During the early afternoon, our short-term guides had become very stretched, which suggested that further upside was unlikely to be sustained, but we also had all the classic signs of a trend day. Those are where stocks open at their lows and close at their highs, with minimal retracements in between.

By 1:00pm EST, the indices had gone on to make higher intraday highs after a gap up open, breadth was strong, all major sectors were participating, and buying interest came in every time the NYSE TICK approached the zero line. When we've seen those conditions last into 1:00, historically the S&P has closed even higher than its early afternoon prices nearly 70% of the time. Since October, it has closed higher 11 out of 14 times.

The last time we've seen a reversal of yesterday's magnitude, after displaying trend-day qualities that late in the day, was July 31st of last year. There was some minor downside follow-through the day following that reversal, then we chopped around. That was a fairly consistent quality of the five other occurrences I was able to find - some minor follow-through the next day, then a choppy move either higher or lower.

I haven't found much about yesterday's reversal bar that would suggest bearish implications. I know it looks ugly on a chart and technicians will be all aflutter about the negative connotations, but I've tested yesterday's bar a dozen different ways, and none of them led to anything that would suggest with a high probability that we should continue dropping here.

We've gone over a handful of different negatives lately, including overbought short-term conditions, extremely low volume, complacent-seeming volatility readings, and a smattering of too-optimistic sentiment readings.

We now also have the Dumb Money Confidence spiking to 71% as of yesterday, the highest reading since January 2007. During the 2003-2007 bull market, such high readings typically coincided with a flattening out of the rally, where we may have seen several more days (or weeks) of choppy upside progress, but those gains were ultimately given back during quick corrections.

During the prior bear market of 2000 - 2002, we didn't see readings this high on many occasions - in fact, it was only seen on August 11, 2000 and January 23, 2001. Neither were good times to be long. I continue to be very leery of upside attempts here and remain modestly bearish in the short- and intermediate-term, but I'll have to re-think that position if the S&P 500 can break and hold over 1420 while the Nasdaq 100 retakes 2000 and the Russell 2000 holds over 740.

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.