Terry, I do not diversify much across industries. But, I do hold a lot of names in my cap gains and 90/10 portfolios. Most tend to be in gaming, biotechnology, info tech, closed end funds, and games, as those are the areas I know best and followed as a professional. But, since I go both long and short, I feel the need for diversification is reduced. For someone who goes only long, I recommend at least 12 stocks in 5 different areas. The 5 areas should be fairly well-removed. For example, you can't buy a pc boxmaker and a software co. and consider that two different areas. Or a biotech co. and a medical devices co. It is only one. A pc boxmaker and an insurance co. would be more to the point. A biotech firm and an Asian closed end fund, also. The 5 areas should also have fairly equal weighting.
True, the rifle approach, one stock you know extremely well, that works out fantastically, will always beat a diversified approach. Just as one stock that doesn't work out will always lose to a diversified approach. I don't think you can take that kind of risk with money you need for your future. But the more names and areas you follow, the more indexed you become, by definition. and the less likely you are to do real well. So, I think my 5 areas and 12 stocks minimum, perhaps 30 stocks, 10 areas maximum, is a good compromise. It gives you a chance to have big winners reward you while it offers some protection against a big loser killing you.
MB |
