Monday, July 28, 2008

Mixed signals

The S&P 500 closed at 1234 today, only 20 points above its bear market low of 1214. The recent bounce peaked at 1282 last week, and the market has fallen 3.7% since then. (I hope that the readers who were considering SSO haven't bought any yet!)

Market indicators are giving mixed signals right now. In the bullish column are a recent trend line break, money flow and the McClellan Oscillator. On the bearish side, sentiment is moderately optimistic. In addition, sector breadth, despite the McClellan Oscillator, is less than ideal due mainly to the opposite motions of the energy and finance sectors.

The market may be setting up for a double bottom, which would require a bounce near S&P 1214 followed by a rally above 1282. If that happens, then a medium-term rally will likely follow. However, until those two criteria are met, we should not assume, hope, or otherwise count on the double bottom occurring; it's also possible that the S&P will plow right through 1214 this week on its way to new lows.

So I'm staying in cash for now, waiting for either a reversal pattern (like a double-bottom) or for more agreement among the various indicators.


Dan said...

Jody -
What do you think of what these people had to say about market timing ?

"Market Timing is a wicked idea. Don't try it --- ever."
[Charles D. Ellis, Winning the Loser's Game]

"I have been following markets for about 50 years, and I’ve never met anybody who could time the market correctly."
(Burton G. Malkiel on 4/28/07 as a guest on “Moneytalk with Bob Brinker.” Malkiel won the 1973 Nobel Prize in Economics for his work on efficient markets presented in his book “A Random Walk Down Wall Street

Thank you for your analysis

John from CO said...

Great info. I for one was considering SSO but have not ventured into SSO yet and will wait for a more clear market direction. Thanks for the reasurrance.

Jody said...

You're welcome, John.
Hi Dan,

I have to give Bob Brinker credit for inviting a timing skeptic on to his show. It takes courage and conviction to face a critic on the air.

Certainly nobody will be able to time the market correctly 100% of the time, particularly given unseen events like terrorist attacks or natural disasters. However, if someone builds a timing track record of 60-40 or 70-30, and beats buy-and-hold over the long run with it, then I would call that successful market timing.

So the real question then is: Is 70% correct timing and/or long-term outperformance possible?

Obviously my experience here will help to answer that question, because I'll either succeed or fail at it. But I wouldn't be trying market-timing if I thought it wasn't possible.

It seems to me that market-timing has actually become easier in recent years, thanks to all of the market data that's available on the internet, and to the phenomenal computing power that I can buy for under $500. (I'm working on my own purely quantitative timing model which I will back-test on the market.) I wouldn't be able to do any of this 15 or 20 years ago, and I imagine there will be even more resources available in the next decade.

I'm convinced that the only serious threat to my market-timing would be if *everyone* started using my method. The whole trick is to see a buying or selling spree coming to and end before the masses realize it, so if everyone were to anticipate it at the same time, then there would be no more advantage left.

Thus, as long as there are skeptics like Ellis and Malkiel, and fundamental investors who insist that earnings drive prices (Brinker), and 401(k) investors who don't even look at their investments for 40 years (many people I know), then I think I have a sufficient advantage in information to use for staying ahead of the market.

hindawg said...

Hi Jody, Curious- what do you use for sentiment indication? isee, cpc?


other John

Jody said...

I use a combination of five different sentiment indicators to guard against any exceptional "bad reading" in one of them.

Dan said...

Thank you for your comments regarding market timing - very interesting and insightful comments to say the least.

I will be paying close attention to your recommendations.


danr said...

How long have you been betting on these medium term swings, and has it beat short-and-hold so far for you (or even been profitable)?

I have come to the same conclusions as you in a few analysis so far, before I read your blog.

So my 2 cents on "timing the market". It's tough to know what the markets going to do next, but it's easy to see what it's been doing up till now. Just use SMA indicators... 10, 50, and 200 work well. They draw an arrow, and we are just trying to stay pointed in the same direction.

It seems to be harder the shorter term you go. My question at the moment is whether to try and trade these medium waves, or whether it would be better to have shorted around 21dec07 (when the 50 crossed below the 200 on SPY), and just hold on through the waves until the next bull market?

Happy days,

ps, the current uptrend seems established now, we passed that 1282 you mentioned.

Jody said...

I'd like to see more than a 2-point gain above 1282 before I declare a trend of "higher highs."

I've been doing my medium-term jumps since the summer of 2007, as described in a previous post.

I agree that using long-term moving averages to gauge the direction of the market (and then go long or short) is a significant improvement over going long with buy-and-hold forever. In fact, just using SMAs (simple moving averages) is probably better than what most market-timers are capable of. It certainly beat Bob Brinker's complicated analysis in the case of this bear market!

The main reason I'm not satisfied with using the SMA crossover method alone is that I can beat it. I seem to have a knack for calling market tops by looking at other indicators, and I'm getting better at calling bottoms, so it makes no sense for me not to use that to my advantage. So far I'm glad I cashed out of SDS (double-short S&P) on the 18th and 22nd, but we'll see what happens. As long as I get back in when the S&P is above 1282, I will have pulled ahead of "shorting-and-holding."

Finally, I agree with you that forecasting becomes more difficult (and pointless) over shorter time frames. I've seen discussion boards where day traders are placing bets based on hourly market movements, and it's painful to watch. I only aim to take advantage of corrections or bear market rallies in excess of 5%, and right now my average position is lasting about 2 months.

danr said...

Thanks for the great answer. That's cool that you can beat it. I hope to learn how to do that.

I also think it's great that you are sharing what you have learned with the world through this blog.


Dan said...

Regarding this recent rally, do you still believe we are in a secular bear market and that this rise over the last few weeks is still just a cyclical bull in a long term bear.

Is this a run that one should go long in say with VTI or should we avoid this trap, if in fact, it is one?

I see that you are still 100% cash.

If you see this:

"I'd like to see more than a 2-point gain above 1282 before I declare a trend of "higher highs."

then would you recommend a long position and if so in what?


Jody said...

My long term forecast can always be found just under my picture on the right side of the blog.

I doubt I will ever take a long position during this bear market. The only reason I'm not staying in a bear fund like SDS for the entire plunge (October 2007 to whenever) is that I can do even better by cashing out at temporary bottoms and getting back in after a small rally; for instance, like March-May of this year.

In previous comment sections some readers have asked how to make money with long positions (like VTI) during these sucker rallies. My advice is to be careful and to use rolling stop-loss orders.

Dan said...

Once again - Thank you!