Let’s assume that at any single time you hold 50 high-beta positions with relatively equal capital allocation of about 2%. Your intention is to hold them for as long as their trend last. You expect some of them will remain your holdings for more than a year and others will get booted in a couple of weeks. When you get stopped in one of those positions, it will be replaced by another. You are constantly 100% invested. When you see signs of distribution in the market, you hedge by buying put options on the indexes or by shorting them. (There will be some very rare periods of extreme weakness when you own less than 50 stocks, because there are no signals in the market.)
You aim for a stop loss of approximately 6% for each position on average. You realize that some of your positions will give you a bigger than 6% loss if it happens that they gap down.
Let’s assume that it is earnings season and two of your holdings miss earnings badly and each drop 25%. What would be the overall damage to your portfolio of 50 stocks? Since each stock has a starting allocation of about 2%, a 25% drop would mean 50 basis points damage to the overall portfolio or 0.5%. In this case, two of your stocks are down 25%, which means that the total damage to your capital is 1%. Since those two have probably violated your stop losses, they are going to be replaced by other two that show promise for future appreciation. Not so bad when it is looked from this perspective, right? Your overall return and good sleep don’t depend on the performance of any one stock.
How do things look on the upside?
Let’s assume that one of your positions triples in a year. What is its impact to the overall capital return. Since each position starts with a 2% allocation, a 200% gain would mean a contribution of 4% to the overall portfolio’s return. Not so impressive when it is looked this way, isn’t it? One stock is not going to make your year.
When your position size is small, a few big winners and a few big losers don’t matter as much to the overall portfolio.
This does not mean that the overall return of your portfolio will be subpar.
Let’s assume that in any given year you have 120 winners and 80 losers. They are more than the 50 positions you constantly hold, because when a stock hits your stop loss, it is sold and replaced by another. The average size of your loser is 6%, the average size of your winner is 20%.
What is your total return?
Since each stock starts with an equal allocation of 2%, the impact of each loser will be (-6%)*2% = 0.12% or 12 basis points. The impact of each winner will be 20%*2% = 0.4% or 40 basis points.
The total impact of all losers will be 80*(-0.12%) = -9.6%
The total impact of all winners will be 120 * (0.4%) = 48%
In this case, the Total capital return is 48%-9.6% = 38.4%
We need to also account for transaction cost in such calculations. The smaller your capital, the bigger the negative impact of transaction cost. Unless you are using a broker like Interactive Brokers, which charges about $1 per transaction or Robin Hood, which promises free transactions or Motif Investing, where you could buy and update whole portfolios for $10.
You don’t need bigger concentration to achieve decent returns. You don’t need to put all your money in two or three names.
Then why some of the best performing money managers in the world use very different, much more concentrated approach?
Soros and Druckenmiller have made names for themselves by taking outsized risks and concentrating big portions of their capital into one or two ideas. When an idea works, they make a ton of money. When it doesn’t work, they lose a lot of money too, but the average size of their big winners far outweighs the average size of their losers. This is a topic I will elaborate on later this week.