It seems you have JavaScript disabled.

Ummm.. Yeah... I'm going to have to ask you to turn Javascript back on... Yeah... Thanks.

Sentiment Overview: Week Of July 2nd, 2010 at Trader’s Narrative

Here is this week’s sentiment recap:

Sentiment Surveys
The weekly AAII survey this week pretty much sums up the sentiment picture we’re going to look at. Retail investors grew rather bold in mid-June, taking the bull ratio to 58%, as the market seemed to recover. But then this week, with the S&P 500 dropping below its recent lows there are very few optimists around.

AAII bull ratio Jul 2010

This week’s AAII shows only 24.7% bullish and 42% bearish. That drops the bull ratio down to 37%, where it was on May 27th 2010 when the S&P 500 was trying to stabilize. Basically, sentiment bounced back much more than the market gave it reason to and now it has fallen back down along with stock prices.

I can’t help but compare the recent correction to the one we saw in November 2009. That decline was just 6% on the S&P 500 but it produced an AAII bull ratio of 28% (meaning 28% of people with an opinion were bullish). The latest decline in contrast has taken the S&P 500 down by 16%. Yet we are still seeing rather elevated bullishness.

Investors Intelligence
Unlike the retail investors, newsletter editors monitored by ChartCraft were basically not impressed with the brief recovery so the bull/bear gap didn’t change much. This week we find them basically where we left them last week: 41.1% bullish and a tiny bit more pessimistic at 33.3%. That give us a bull/bear ratio of 1.23:1 not very different from last week’s 1.32:1 and not even that much far from early June 1.21:1.

Secret Survey
I’ve recently stumbled on a sentiment survey which is not very popular at all but which has all the makings of an amazing sentiment indicator. It is weekly, it targets retail investors, and it goes back several years. A cursory study of the available data shows it to be a contrarian indicator. Because I’m not finished researching it, I don’t want to reveal too much but I’m too excited to keep totally silent.

Right now, according to this survey there are 0.57 bulls for every 1 bear. That’s quite a bit of pessimism but it isn’t extreme. There was an extreme reading in late May which was remarkably lower than the reading from March 2009! I hope to have a better understanding of it in a few weeks at which time I’ll write more about it.

NAAIM Survey of Manager Sentiment
Active asset managers surveyed by NAAIM suddenly changed their minds and significantly reduced their exposure to the stock market this week. The most recent low for this indicator was 27.05% in early June just as the market seemed to finally have found firmer footing. As a result, the NAAIM sentiment sharply increased to 54.24% last week.

NAAIM survey of manager sentiment Jul 2010

As I noted in last week’s sentiment overview, this was surprising and bearish since the market had actually fallen slightly. On June 2nd, the S&P 500 stood at 1102.83 and on June 30th, it closed at 1092.04. But the (mean) NAAIM sentiment had increased during that time from 27% to 54%.

So what we are seeing this week isn’t all that surprising when taken in that light. Stepping back further, the NAAIM sentiment is still rather elevated at 36% - being higher than where it was on June 2nd - even though the market is now 75 points lower. This means that there is a bit of insistence on the part of asset managers as the “look on the bright side of life”. If you’re looking for a lasting low, as a contrarian you would much rather see them throw in the towel with disgust. Since that is not what they are doing now, previous patterns would suggest that we need lower prices to induce such a capitulation.

Consumer Confidence
The Consumer Confidence index (from the Conference Board) fell to 52.9 from the revised level of 62.7 in May. The consensus expectation was a slight decline (62.5) but the actual number produced the largest decline since February 2010 and took the index down to a 3 month low.

Historically, during recessions the Consumer Confidence index averages 70 and during economic expansions, 100. So according to the continued low readings in this index, the US economy is still mired in a deep recession. Other parts of the survey are equally pessimistic. Only 10% of respondents responded that they believed business conditions were good.

Mutual Fund Flows
US retail investors continue to withdraw money from their equity funds. According to the latest ICI data, domestic equity funds have seen an outflow of $7.9 billion for the month of June. This on top of the $23.6 billion that was taken out in May 2010.

Bond funds continue to be the favorite garnering the majority of the money being socked away. But even this investment destination has seen a drastic reduction. Last month bond funds received $19 billion and in May they received inflows of $16.9 billion. This is much lower than the multiple months from July 2009 to March 2010 where bond funds had inflows of $26 to $47 billion.

Rydex Traders Sentiment
The market has done the same “head fake” to Rydex market timers. The assets of the S&P 500 index inverse fund (previously known as URSA) surged as the market declined and then receded with the short lived recovery. They have now increased once again and are very close to hitting the levels of last summer:

Rydex URSA inverse fund assets compared to SPX Jul 2010
Source: McClellan Financial

While the nominal asset levels are contrarian bullish, what concerns me is how ready the Rydex traders were to believe in it.

Corporate Buybacks
According to Standard & Poor’s US corporations increased their stock buybacks by almost 79.5% in the first quarter of 2010 to $55.3 billion, compared to last year’s first quarter. This is the 3rd consecutive quarter that S&P 500 index constituents have increased their buyback activity.

It seems that at least some of the cash hoard on companies balance sheets is being directed to share buybacks. Among the sectors, the technology sector which has been leading the charge in performance on the index accounts for the largest share of buybacks (29.3%). Consumer staples is second with 19.3% of the share of buybacks. In contrast, utilities and the telecom sector actually decreased their first quarter buyback programs.

Option Sentiment
The lack of concern from the options market should be noted by the bulls. While the S&P 500 index has clearly fallen below the late May lows, the CBOE put call ratio (equity only) is not showing any real sign of concern. In fact, the 10 day moving average which tracks the past 2 weeks of trading for puts and calls is now lower than it was in late May.

On Tuesday, when the S&P 500 decisively broke below the lows in May, the CBOE equity only put call ratio was shockingly subdued at 0.79. On a day when major support is broken and the index closes down more than 3% you would imagine that we at least see the ratio hit 0.90 or even 1.0!

The same complacency can be seen on the ISE with the ISE sentiment index falling to 126 on Tuesday (implying that option traders were buying 126 calls for every 100 puts). That is not indicative of any real “fear” or even concern. Even more surprising, on Thursday when the market was weak - but closed higher than the intraday lows - the ISE equity only call put ratio was 200!

Gold Sentiment
As soon as I reduce my skepticism regarding gold’s relentless bull market it drops more than $40/oz. or 3.52%. The DSI for gold is still elevated at 90% but the Hulbert Gold Newsletter Sentiment index which tracks the small group of stock newsletter editors that try to time the precious metal shows a different story.

According to the HGNSI not only is there a total lack of excitement about gold’s performance, especially since it is so close to its high, but the recent drop has caused even more advisors to jettison it from their holdings.

In late June when it was creeping higher close to taking out the $1250 level, the HGNSI was at 37.8%. Now, after Thursday’s drop, it is 23.5%. And at the top in early may it was only 46.6%! I’m not sure what it will take to get the gold bugs excited about the yellow metal again. Over the years, a HGNSI reading of ~70% has corresponded with tops. Since we are very far away from that now and since there is no stubbornness shown on the part of gold bugs, this would suggest that gold can recover and continue on its way.

Similar to the HGNSI, Rydex gold market timers have also throw in the towel. As the fund itself has lost 10%, from June 25th, so has the asset base shrank 10%. Also consider the fact that the assets for the Rydex precious metal fund are still very low compared to the last time that gold was trading this high (December 2009). This tells me that we aren’t seeing a rush of speculative after gold and that the trend could continue.

Enjoyed this? Don't miss the next one, grab the feed  or 

                               subscribe through email:  

13 Responses to “Sentiment Overview: Week Of July 2nd, 2010”  

  1. 1 bill

    no need for the secrecy. the sentiment survey you are referring to is the TSP talk survey and it very well know among the trader community as many like myself have been watching it for several years now.

  2. 2 Babak

    Bill, that teaches me for misunderestimating my readers… lol
    what else have you guys been keeping from me, eh?

  3. 3 david

    check out

  4. 4 jeff

    Any way to take all of your sentiment indicators and enter them into an excel spreadsheet w/ values to determine a complete bullish/bearish bias?

  5. 5 Aly S.

    That’s a good idea, jeff.

    Something like a sentiment composite index!

  6. 6 Wes


    Increased volatility has created extreme price levels in the past few years, and because of this I believe many “public” speculators and investors have simply quit the market. This could lead to somewhat suspect sentiment indications generated by the surviving traders.

    One group I follow closely that has not changed in make-up is the group composed of traders of the highly liquid OEX options. This group is dominated by the big smart professional traders who have an impressive record of being correct about market direction, buying calls in front of major market increases and puts before market declines.

    The OEX options traders have been consistently bullish during the current correction, and their behavior during the last 10 days, which have all been down days for the OEX 100, is especially noteworthy. During this 10 day period there has only been a single day where their purchases of puts exceeded their purchases of calls (June 22 @160 P/100C). Even this single day was not particularly out of line high, as they frequently purchase twice as many puts as calls.

    During this 10 day period there have been two exceptionally bullish readings of 39 and 42 on Monday(the 28th) and Tuesday of this past week. All other readings during this 2 week period have also been bullish, and the average P/C for the period has been 81.6, solidly below the bullish level of 95 and almost at the very bullish level of 80.

    I also look at the open interest of the puts and calls generated by these traders to make sure that day trades are not distorting the figures. During the 2 week period in question their OEX P/C open interest has averaged a solidly bullish 89.5. Historically, the open interest P/C bullish level is 120 and the very bullish level is 100. The extremely bullish level, hit consistently in early 2009, is 80.

    So, there seems to be no question in the collective mind of these successful options traders that we are in a simple correction, soon to be followed by more upside.

    I have checked the historical record (all this century) and cannot find where they have been wrong before. This could be the first time, but I’m not going to bet against them.

  7. 7 Wes


    As an addendum to my OEX option traders note, I just noticed that in the past 5 days, which were easily the most bearish looking during this entire correction, the smart OEX option traders averaged an exceptionally bullish 61 puts purchased per 100 calls.

    I haven’t checked for records, but this 5 day period must rank amongst the most bullish.

  8. 8 Wayne

    A comment on trying to develope sentiment composite indexes.

    I have always wondered why we have the American Music Awards and allow the public to vote for the best music, when they have already cast their real ballots at the record or CD. Which is more important, how they vote in a poll or how they vote with their money?

    The same goes for sentiment composite measures. Every time someone buys or sells a stock, they cast their real vote. If the market is down 20% over the last quarter, then a lot more investors were bearish than bullish over the last three months.

    So simply taking every poll you can get your hands on and adding them together wouldn’t add a lot of value in my opinion. I would think that the better approach is to concentrate on those particular sectors of investors that have a long history of either being right or wrong at major turns in the market. I

    Although I read “The Art of Contrary Investing” twenty years ago, I don’t profess to be an expert in the area, and don’t have any personal research to support what are merely observations, but it seems as though looking at options trading by institutions vs individuals would makes sense, just as insider trading vs odd lot trades, etc, possibly AAII polls since they represent the retail investor, who normally the last to show up for the party. Hulbert should take a ratio of their top 10 ranked newletters vs their bottom 10 newsletters.

    Any measures of smart money vs historically wrong money. As is the case with most market models, less indicators or components is probably more.

  9. 9 bill

    I have personally spent several years trying to develop mechanical trading systems based on various sentiment indicators.

    The biggest problem with such an approach is that it doesn’t take into account the changing context of the market. The buy/sell signals from sentiment indicators occur at different levels depending on whether one is in a bull or a bear market trend and depending on the strength of that trend.

    For example, the put/call ratios levels that were a buy signal in the late 90’s bubble years would be sell signals in today’s market. And you can see that same thing with just about any sentiment indicators that reached major extremes of bearishness during the 2008 bear and yet produced no rallies, and the opposite was seen during the 2009 melt-up where sentiment stayed overly bullish for long periods of time without producing any meaningful corrections.

    The stock market is an extremely complex and dynamic system so simple linear buy/sell rules based on any sentiment based indicators with a limited historical sample size are not robust and are bound to fail. Sentiment is only one piece of the puzzle. Liquidity and Monetary Policy is another very huge piece of the puzzle and can easily override sentiment extremes in my experience. And of course valuations/earnings trends are another major factor.

    I think one has to look at all of those factors and not just sentiment in a vacuum in order to successfully time the market.

  10. 10 jeff


    For those same reasons it explains why backtesting often yields unrealistic results. Running a bullish scan in a bear market will not have positive results.

    While I agree with you that it is difficult to use sentiment to time the market there is a lot of value in using them for extreme situations, regardless of what type of market we’re in.

  11. 11 OntheMoney

    Wes -

    Take another look at the OEX options indicators. They are, as you say, a non-contrary indicator and usually an excellent guide. But they have indeed been wrong before.

    I’m looking at SentimenTrader’s charts of their activity over the last decade. From November 2007 to March 2008 they were almost permanently bullish as the market slid over from bull to bear.

    You could, however, have discounted the bullish signal by looking at the open interest at the time, which was bearish. Yet look at the same measure from late 2000 to late 2002. All the way down, open interest was full-on bullish.

    It’s usually at the turning points that these indicators can fool you, especially if you’re looking for a particular bullish/bearish bias to be validated.

  12. 12 bill


    I agree with your points, but I would add that when these sentiment indicators “fail” it is not so much that the indicators has stopped working but that other stronger forces have overridden it’s bull or bearish effect. For example, if stocks are overvalued and/or earnings trends are decelerating sharply then bearish sentiment extremes may not be enough to produce a meaningful bounce and of course vice versa for bullish sentiment extremes. Same goes for liquidity and monetary conditions or solvency issues which can cause either forced selling via deleveraging or cause bubbles when leverage is expanding rapidly either of which will be fairly immune to being reversed by sentiment extremes.

    Sentiment extremes are but one piece of a very large complex puzzle. In some of the studies I have done the largest correlation I have ever found between the market and any sentiment indicator was around 30%, which would indicate that 70% of market action is caused by factors other than sentiment. I think Hulbert has shared similar numbers.

  13. 13 AB

    Aside from assuming that an extreme in one end of sentiment is a valid contrarian signal, one should also look at price extremes - I don’t recall anyone here mentioning moving average spreads being extreme, and if they did, they’d know that prices on the S&P 500 were at a persistent bullish extreme (4%-5% for my spread) in late 2009 and then April 2010, and that historically, the index tends to back off pretty hard from that level. And the fact that there was an extreme in the total put-call ratio on April 14 - meaning that no one was hedging on the downside - gave sellers extra fuel by which to torch the complacent bulls.

    Also, selloffs always require catalysts of some sort, and April had Goldman Sachs vs. SEC and then the Transocean/BP incident, and then the disturbance in Greece.

    About the $OEX options, just because seasoned traders are making bullish bets, that does not mean an automatic turnaround - in fact, they could well be doing the same thing as creating a ‘wall of worry’, except to the down side. Any time there’s a conflict in sentiment, then the dominant trend is being given the fuel (trading against the trend) to continue.

    The bright side is that the 10-day moving average of the total put-call ratio formed a much weaker bearish extreme, which is perhaps what those $OEX traders are playing.

Leave a Reply