Winning Strategies

[DOTTT]

I hope we can get some good ideas in this thread. If you any investment strategy that is profitable please post it here. I'll get the ball rolling.

The 4% Rule

I read about the 4% rule in Martin Zweig's Winning on Wall Street. It is part of Zweig's "the trend is your friend" theory. The 4% rule can be used with any market index or mutual fund you like. All you need to construct the model is the weekly closing price for your chosen investment vehicle.

A buy signal is given on any weekly close 4% (or higher) above any recent weekly closing low.

A sell signal is given and the money can be moved to an interest bearing account on any weekly close 4% (or more) below any recent weekly closing high.

From 1966 to 1988 Zweig's 4% rule, using the Value Line index, returned 14.9% per year, versus a buy and hold of 2.4% per year. The hit rate was 52%, with being on the right side of all the big moves responsible for the big gains.

You can be as aggressive as you like with this strategy. You can trade the QQQQ, use leverage, or trade futures. I would love if someone can run the numbers on how the rule would perform on the sell side.

I have a couple more strategies I'll share later.

[gull]

Why did you choose 1966 to 1988?

What are the returns for, say, 1928 to 2006 in the major domestic indicies?

Please provide evidence to back up your claim (I don't mean this in a rude way. Just link us to the study).

[Big TR]

Cherry-picking this 22 year period seems a bit strange.

I too would be interested to see what this method returned for the most recent 22 year period from 1984 to 2006.

This seems like a "Dogs of the Dow" type of investment strategy, where if you look at the same data dozens of different ways, you will find some common factor that would beat the pants off the market for that given time period.

[DOTTT]

I didn't run the numbers myself, I think these were published in Martin's book and the reason why he selected that time period specifically is to highlight how terrible a buy and hold strategy can perform. I'm currently working on running the numbers from 1929 to 2006 and will let you know the results as soon as I finish them.

[DOTTT]

Interest Rate/Moving Average/ NASDAQ strategy

This next strategy proves using interest rates as buy or sell indicator can lead to market beating returns.

This strategy is very simple to use, a buy signal is given when the FED lowers interest rate, and the NASDAQ is above, or crosses above a 22 week moving average. When the FED begins to raise interest rates, the NASDAQ position will be sold and the money moved into a money market fund.

That simple strategy resulted in a 5.7% annual return higher than the S&P500, and for 28 years! And with half the risk, since you are out of the market approximately 50% of the time.


1973 to 2001......annual return.....10K becomes
--------------------------------------------------

Nasdaq/interest/
moving average .......14.1%.........$419,832

Nasdaq buy/hold.......10.8%.........$174,019

S&P500 buy/hold........8.4%..........$99,999

[DesertCat]

[ QUOTE ]

From 1966 to 1988 Zweig's 4% rule, using the Value Line index, returned 14.9% per year, versus a buy and hold of 2.4% per year.

[/ QUOTE ]

I highly doubt the buy and hold return was as low as he claims. A common error when making these comparisons is to use an index (such as the Dow) and simply calculate it's gain over the period from the beginning value to the end value. This ignores dividends, which add anywhere from 2% to 5% per year to the buy and hold approach. Dividends used to be a much bigger component of stockholders gains, it's only been the last twenty years or so that dividend rates have declined probably due to the lower capital gains taxes.

And Zweig's rules don't appear to be eternal. From the Amazon reviews of his book, here is one critique.

[ QUOTE ]

Zweig made his reputation as a market timer, and 2/3 of this book presents a detailed market timing model that incorporates both "money indicators" (e.g. prime rate, fed funds rate, consumer debt) and a basic momentum indicator. The model is relatively simple and the method is clearly explained. According to Zweig's data, the system produced remarkable results up through the final revision of this book (in 1996).

But, of course, you have to wonder... The book has been revised four times since its initial publication in 1986...and yet not a single revision in the past ten years. Hmmmm...wonder why?

Unfortunately, the obvious answer is the correct one. Zweig's "Super Model", which he touts as "The Only Investment Model You Will Ever Need" (yes, that's an actual chapter title), utterly failed after 1996. Some other reviewers claim to have followed the model successfully since the last edition of the book. I don't know what numbers they're working with, but I've done the very tedious work to recreate the signals the model would have given since 1996, using only actual data available as of the date it became available, following Zweig's methodology precisely, and applying it to the Value Line Arithmetic Index (a very close substitute for the proprietary benchmark he uses in the book). From March 1996 (the last data point in the book) through June 28, 2006, the Value Line Index produced a gain of 222.8% (buy & hold, excluding dividends). Following the Zweig "Super Model" long-only generated a gain of 95.5% (not including interest income while in cash), and following the model long/short produced a gain of only 18.4% (yes, that's total, not annualized...). So much for the "Super Model".

As for Zweig's stock picking method, it's a pretty straightforward approach blending GARP and momentum and is very capably summarized on AAII's excellent web site. Save yourself the time and money and just go there if you want a starting point for stock screening ideas.


[/ QUOTE ]

Zweig's formula isn't unique, the world is full of junk "formulas" that have been extensively backtested and shown to work over the last ten years, twenty years, etc. The problem is you don't need a formula that worked in the past, you need one that will work in the future. Extensive academic research indicates that historical patterns are as likely to continue in the future as flipping a coin that's come up heads thrice in a row will keep coming up heads. For example, momentum investing was hugely successful in the late 90s (as it will be in any manic up market). When the bubble popped, it was the fastest way to burn your portfolio to the ground ever invented.

The good news is there are investing approaches that work and have withstood the test of time. Warren Buffett learned his approach from Ben Graham, who started using it in the thirties. Buffett's been successful at it for 50 years. More successful than Zweig, or pretty much anyone else.

[Slowroller13]

[ QUOTE ]
Buffett's been successful at it for 50 years. More successful than Zweig, or pretty much anyone else.

[/ QUOTE ]

O RLY?

List of AAII Stock Screens

25% compounded annually over 19 years

Am I knocking Buffett? Absolutely not. Ben Graham's approach has withstood the test of time.

I just wanted to point out that Zweig is no slouch himself.

[pig4bill]

Times, they be a changin'. What's the copyright on that book? The market is much more volatile than when Marty probably wrote that book. 4% of a $30 stock is $1.20. A lot of $30 stocks will have that wide a range in a day, much less a week.

BTW, he was one of my favorite guys to watch on Wall Street Week.

[DesertCat]

[ QUOTE ]


O RLY?

List of AAII Stock Screens


[/ QUOTE ]

I have no clue why a list of AAII stock screens is pertinant here. You are just seeing the results of self selection, formulas that best match history get ranked highest, but that doesn't mean they'll be successful in the future. And the historical results aren't necessarily accurate either. Does the AAII system incorporate trading costs, taxes, and management fees?

Warren Buffett doesn't use a mechanical formula (no great investor does), so no matter how much the AAII and Hagstrom tries, those screens aren't representative of Buffett's approach.

[ QUOTE ]

25% compounded annually over 19 years

25% over 19 years: Zweig Performance Ratings Category 1 stocks from May 1976 to March 1995, achieved a return of 6,793%.


[/ QUOTE ]

Well from 1975 to 1995, Berkshire Hathaway grew investments per share from $159 to $21,817. that's an increase of 13,621% and it's AFTER all taxes and fees and all transaction costs. These are real world audited results, not the on paper result of a backtested formula, or claimed results by a secretive investment operation.

And Buffett's results were even better before 1975, and are still pretty good after 1995. His job has gotten ever more difficult as Berkshire's capital has grown, which makes his results even more stunning (right now he's responsible for investing over $100B).

I don't know what audited results Zweig actually produced. I don't know how much money he actually managed. It's orders of magnitude easier to beat the market managing $1M than it is managing $10B. I don't know why he stops claiming results after 1995. I don't know why he picked 1976 as a start date. Buffett's yearly results from 1976-1985 were 59.3%, 31.9%, 24.0%, 35.7%, 19.3%, 31.4%, 40.0%, 32.3%, 13.6%, and 48.2%. It was a good time to be picking stocks.

Before you fall in love with Zwiegs formulas, you should have good answers to all those questions. He oversees the Zweig Fund which has an annualized return of 7.6% per year since 1986, and a negative 2% per year return over the last 5 years. So his approach is so good, why is his real world fund having such a horrible run over the last two decades?

[DOTTT]

Here are the results of the 4% rule tested on the Value Line index. I originally wanted to run this on the S&P 500 but was unable to get the data. The 4% system gained 10.9% a year from 1961-2006, but get this drawdown was held to a minimal 13%. ValueLine buy and hold returned 3.3% a year with a horrific 76% drawdown.