Trader’s Narrative

Sentiment Surveys
No significant change within the usual sentiment surveys. Both the AAII and II moved just 1% points: the AII going to 47% bulls from 46% and the Investor’s Intelligence bullish percent from 45% to 46%.

Hulbert Newsletter Sentiment
According to Mark Hulbert, out of 10 market timing newsletters with the best risk adjusted performance, seven are bullish right now. Among the ten, their average equity allocation is a whopping 78%! This is especially meaningful when compared with what the worst market timers are saying: an equity allocation of just 11%.

So unless you think that all of a sudden the pattern of performance is going to be upended, this is a very bullish sign for the market.

ISE Sentiment
We got a surprising uptick in the ISE Sentiment index this past week. On Thursday it peaked at 192. To provide some perspective, on October 8th, 2008 the ISE sentiment index reached 279 and on October 29th 2008 it reached 275. The October swing high of around 1560 was put in between these two dates.

And finally in mid May, when the market made its swing top, the ISE sentiment was 225. We didn’t really approach record highs for this measure but usually anything around 200 shows significant risk aversion (the ISEE sentiment is a ratio of calls to puts).

LowRisk Sentiment
A few people contacted me asking about LowRisk and as I mentioned before I haven’t received any response from Jeff Walker after inquiring about LowRisk. Their site continues to be stuck on March data with no updates as of yet. Unfortunately, it looks like this sentiment measure is no more.

So more or less we continue to muddle through with lukewarm sentiment on average. I continue to see the March bottom however, as significant and these counter currents we’ve seen lately don’t necessarily negate that.

LowRisk is probably one of the least known and followed investor sentiment surveys.

It is a bit of a mystery how large the sample size is because Jeff Walker, the person who runs it, won’t disclose how many respondents he gets or has gotten historically. In any case, all we can do is take the data and see if it provides an edge or not.

Every week, anyone with a computer and internet access can fill out a simple question:

How do you think the DJIA will perform in the next 30 days?

• Up 2% or more
• Flat (less than 2% up or down)
• Down 2% or more

There’s also another metric where they ask you to “guess the Dow”. LowRisk has been collected non-stop since 1997 and has proven to be a good contrarian measure.

Which is why it is notable that for this week it flipped from a neutral reading to a very bullish one. Right before the recent market swoon (June 3rd 2007), LowRisk showed 40% bulls and 34% bears.

But on June 10th 2007, there were 14% bulls and 60% bears.

That is very extreme. Even for LowRisk which is one of the most volatile sentiment surveys. In fact, there were only a handful of times in the past with such few bulls:

March 27 2005 —- 14.50%
July 25 2004 —- 14.30%
August 21 2005 —- 12.10%
April 18 1999 —- 12.00%
October 17 2004 —– 10.50%
September 19 1999 —- 10.30%
September 9 2001 —- 10.30%

During those past instances it did a spotty job of pinpointing great buying opportunities. Almost all of those dates correspond to a small bounce, but they were usually followed by a much better buying opportunity (lower bottom) in the immediate future.

There is another way we can slice and dice the data. Lets look at the bull ratio: bulls divided by the bulls plus the bears. This way we compare the two extremes and take out the neutral bystanders.

By that measure, the latest bull ratio of 18.9% is also one of the lowest historical instances. Here are the other times that the bull ratio was lower than 20%:

September 2 2001 —- 12.20%
September 19 1999 —- 13.84%
October 17 2004 —- 15.35%
August 21 2005 —- 16.64%
April 18 1999 —- 17.65%
March 9 2003 —- 18.06%
March 18 2001 —- 18.20%
September 25 2005 —- 19.37%
March 27 2005 —- 19.46%

As you’ve no doubt noticed, there’s a lot of overlap between the two lists. So what I said above applies to this list as well. The only standout date, because it turned out to be an amazing time to buy, was March 9th 2003. This was the third and final retest of the bear market bottom, just before the lift off to a new bull market.

Now the caveat. This is the most bearish (therefore bullish by contrarian analysis) sentiment measure. But it is just one data point.

Also, as I’ve mentioned before, LowRisk is very skittish and jumpy. Which is why it is usually smoothed with a short term moving average. Right now, its moving average is still in neutral… although with a few more numbers close to this extreme low and we’re going to see even the moving average reach extremes.