July 3, 2007
Book Review: How I Made 2,000,000 in the Stock Market by Nicolas Darvas
Nicolas Darvas wrote a very entertaining book. Maybe the fact that I'm half Hungarian added to the vicarious enjoyment. His early misadventures touched a raw nerve. His remarks about the high cost of trading in the bad old days reminded me of why I switched to an online broker. While reading the book I broke out in laughter several times.
I was so intrigued by the box model that I created a computer model to test it. The results were a disappointment. While I didn't test it with too many stocks, I ran the tests with growers like BLWD, AAPL, and CMG. Darvas stated that he invested in growth stocks even if he didn't call them that and I tried to mimic his style as closely as possible. To give an example, over a period of 18 months from January 2006 to June 2007, the model traded BWLD five times, two winners and three losers for a net gain of 11% in 18 months. Improving the stops might have increased the yield to around 16%. That's nothing to write home about. During that period of time the Dow 30 advanced 16% (annualized percentages).
Some seven years ago I played around with another computer model based on crossing the various DMA lines. The stocks I used were the tech darlings of the bubble era. I used real price data from the '90s. The model made Warren Buffet look like a rank amateur, it was producing yields well in excess of 50% and it generated a very concentrated portfolio even when I fed it data for over 100 stocks. When I tested the same model in a bear market scenario it came out even, losing the trade commissions.
Why the long story? Because I'm trying to put Darvas' achievement in perspective. From Jan 1957 to Dec 1958 the Dow 30 yielded 24% annualized, about three times its long term average. Darvas used an enormous amount of leverage or margin, more than most of us can obtain today and certainly much more than I would ever be willing to risk. Darvas must have had very good income from his dancing to dare risk that much. He did keep is income and his investing funds separate. Darvas was also very astute in his stock picking. The Soviets launched Sputnik on October 4, 1957 and America was feverishly trying to catch up. Kennedy was a good speechmaker.
I believe that an extraordinary number of favorable circumstances came together in a very fortuitous way to enable Darvas to achieve his feat. From $25K to $2.25M in two years is an annualized return of 843%. As they say on TV, "Don't try this at home." :)
Some 30 years later a similar set of circumstances came together again, a bull market and a technology bubble. In the 17 months from September 1998 to February 2000 my portfolio more than tripled with hardly any use of margin at all.
The concept of "Expensive-but-cheap, high velocity stock" is perfectly clear, it's a high P/E, fast grower. In the '90s it would have been AOL, CSCO, EMC, MSFT, DELL and a few others. See baggerness 12/19/1999. Of course, by the time I wrote that essay the bull run was reaching its end. Anyone who latched onto these stock in the '90s could have repeated Darvas' feat. The difficulty, of course, is finding the current "Expensive-but-cheap, high velocity stocks" and having the guts to invest in them with margin. I don't think this kind of stock shows up too often, it usually goes with a "New Paradigm Shift." Solid state electronics sure was that in the late 1950s and early 1960s. Then it morphed to PCs in the '80s. Then it morphed to communications in the 1990s. The next one is probably something that feeds on the Internet. iPod, iTunes, iPhone?
I was able to build what I think is a fair reconstruction of the box model but I admit that the Q&A addendum helped. As for mental charts, I have known people who could do stuff in their heads that ordinary mortals have a hard time doing. My dad, being in the hotel business, could tally up a sum almost as fast as someone with an adding machine and he could reel off the correct exchange for the commonly accepted currencies at the hotel. Auto parts dealers know the number of thousands of parts without the need for a catalog. It comes with the territory. It's important to remember that back in the 1950s we didn't yet have all the wonderful metal labor saving machines we have today. You still had to use your brain for many tasks that we now routinely assign to machines such as spreadsheets.
Believe it or not, Warren Buffett and Nicolas Darvas have one thing in common, both used an inordinate amount of leverage. The difference is that Darvas was trading on margin, a very risky strategy, while Buffett uses other people's money and float, a much less risky strategy. Buffett started with about $100 in his initial partnership while his partners invested $105,000. Later he used Berkshire Hathaway's working capital. Still later he graduated to using the float from insurance premiums, and trading coupons. When Buffett started to buy entire businesses, he used their excess working capital as a source of investing funds. Leverage to the max but with very little risk compared to broker supplied margin and at a much lower cost.
I'm very much tempted to use the box method but I don't think it will work except in a strong bull market and then only if I can identify the "Expensive-but-cheap, high velocity stocks." This is no easy feat because the high P/E ratios frighten me. But I think there is an alternative that can also produce excellent results which is to use long term LEAP options. Options have leverage built in without having to pay extra interest charges on borrowed money. If you can identify the stocks that are in their own secular bull runs then the corresponding LEAPs should produce excellent results. I stared to experiment with LEAPs just six months ago and, so far, the results have been very satisfactory. That said, I doubt I'll ever write a book with a title similar to Darvas' book :-)
Copyright © Software Times, 2000, 2001, 2003. All rights reserved