Funny situation, this. I took a Bear-o-meter reading today (Wed. June 13). The indicators read as follows:
- Primary & secondary trend (200 day, 50 day SMA): Bullish
- Breadth (AD line, Dow Theory): Bullish
- Breadth momentum (New high/low, % stocks over 50 day SMA): Neutral
- PE ratio: Neutral
- Sentiment (Smart/Dumb, VIX, Put/Call): Bearish—with one possible exception per below
- Seasonality: Neutral
After the above factors were tallied, the Bear-o-meter came out to either a 2 or a 3. Before sending me hate-mail, let me explain what I mean by such a wishy-washy reading on what is supposed to be a quantitative indicator.
All of the above factors gave clear signals that put them firmly into bull, bear or neutral ratings. The exception to that was the “Smart money/ Dumb money” ratio. It is, as of Tuesdays close, a mere 0.04% over the trigger level to put it in a bearish position. So you have 2 choices:
- If you want to assume it will move into a bearish position by Thursday or Friday, you can give the Bear-o-meter a “2”—which is firmly in HIGHER RISK territory.
- If you would rather say that you will NOT assign a negative score until it officially breaks –then the Bear-o-meter reads “3”. That level implies a marginal risk score.
Put to Call: Bearish without a doubt!
Smart/dumb money spread (courtesy Sentimentrader): On the verge of a bearish reading….
Sliding cash scale
One of the changes we brought into our model earlier this year was to assign a sliding scale for cash holdings within the Equity Platform. We prefer to quantify our decisions wherever possible, and this sliding scale offers a more quantitative tool to control the cash level in our equity model. Our new sliding scale cash model has assigned a MAXIMUM 20% cash weighting assuming a score of “3”. If the “2” rating holds, it would move us to a MINIMUM of 20% cash, with discretion up to 40% cash!
Conclusion
Given the the sentiment indicators are vastly faster moving than some of the other indicators I use in the Bear-o-meter model – I’m inclined to think that if there is risk in the markets – we may only get a quick pullback in the nearterm, and then return to a firm 3-4 “neutral” reading. Should a quick pullback be in the cards, I would likely spend about half of my harbored cash within the model.
Hope this has given you some food for thought!
5 Comments
Thanks Keith. I’d like to clarify.
Does this mean that the dumb money (people like me) is what’s driving this rally?
Also, with regards to the modest bear-o-meter reading, you’re thinking that it’s a good idea to have +/- 20% cash and then invest 1/2 of it at the next pullback. I assume pullback is about 5 or 6%?
Hi Sally
Yes, it has been a retail driven market of late, aka dumb money. The smart/dumb confidence ratio I show doesn’t split them apart–but I sometimes post the two lines separately rather than as a ratio–and on that chart you can clearly see just how enthusiastic the retail money has been on this market–meanwhile, institutions have been hedging.
Re the cash: We are trying to be more quantitative with our cash allocations. So our model now dictates somewhere around 20% cash–thus we hold that amount. So I guess you could say I do feel its a good idea to hold cash–but its less about an instinct/feelings, and more about the relative risk on the markets and how I should allocate cash according to that risk/reward status – This is all about the indicators that comprise the Bear-o-meter compilation. Its largely neutral or bearish in most of the short termed components (sentiment, breadth momentum) while its bullish in the longer termed components (trend, cumulative breadth).Thus my thoughts that we may see a neartermed soft spot – but a longer termed uptrend. so–I am likely to buy on a dip.
Keith, I appreciate your recent generous sharing of the factors that go into your Bear-O-Meter. I wondered if I should use it to help with my decisions. But two things kept/keep me from doing that.
1. The assignment of “points” to the various indicators seems arbitrary. Admittedly, I may not understand how you arrived at these values in order to give greater weight to some factors over others.
2. This must have worked well for you over the long term, otherwise why would you use it? But recently, it has given wrong signals. The market has continued upwards and we all know that the market can continue to be irrational longer than we can remain solvent. And if we don’t make money when the market is going up, when are we going to make money? (Assuming, long only positions.)
So, I wonder if you may need to go back to the drawing board and figure out why your Bear-O-Meter is not working in the current market conditions.
With respect and thanks for allowing us to comment.
Hi Paula
The Bear-o-meter is not so much of a timing instrument as it is a risk/reward assessment. Thus, it gives a probability of risk (or reward) going forward, and the differnet ratings reflect that probability.
As far as assigning points – I must admit my weighting of the factors comes from 30 years of doing this stuff, and not from a computerized backtest. Eg: I am very aware that a break in the 200 day SMA is quite important -thus the markets position relative to it gives the Bear-o-meter a 2 point assignment either positive or negative. Less important is a break in the 50 day SMA–thus a 0 or 1 point assignment. The 50 day SMA doesn’t get a negative assignment, given that a break of that SMA to the downside occurs frequently within a bull market. As such, it simply gets a 0 if the market moves below it.
That’s just one example, but what I want you to understand is that each factor has been assigned a point or 0 or negative according to my personal observations during almost 30 years of experience. I have used these factors long enough to get a handle on how significant each factor is. But its not quantitative – to your point – except that my experience has taught me that certain things are more significant than others in observing market risk and trends. I assign the weightings accordingly.
As far as accuracy–as you note, its forward looking. However–it has not given bad signals at all recently It called the January change in risk on Jan. 19th-where t went from a prior “5” to “3”-very near perfectly catching the top, and then a re-entry point with an “8” score on Feb 9th–again, almost perfectly catching the bottom.
In fact—This blog highlights a possible bearish change. As I type this reply to your question, the market has moved down 2 days in a row….so…I’m not disappointed in its signals.
Again, though, I don’t get too caught up in using it for timing, and neither should you. Its more about RISK. A low reading simply means you are not operating in as low a risk environment as would be ideal. But that does not mean automatic selloff by any means.
Hope that helps.
Keith,
Your reply was helpful: risk/reward assessment, NOT a timing tool. Got it! Thanks!