One of the dangers about net-net investing is that if you buy a net-net that begins to lose money your net-net goes down and your capacity to be able to make a profit becomes less secure. So the trick is not necessarily to predict what the earnings are going to be but to have a clear conviction that the company isn't going bust and that your margin of safety will remain intact over time.

Long term debt and bank debt (including off-balance sheet financing must be judiciously employed. There must be room to expand the debt position if required.

To put money into anything, anywhere, provided that the downside is measurable and acceptable and the chances of a good profit appear to be better than 50%. I will not take gambles, but it is part of my job description to be ready to take very carefully calculated risks.

When a stock doubles, sell half - then what you have is a free position. Then it becomes more of an art form. When you sell depends on individual circumstances.

To my knowledge there are no good records that have been built by institutions run by committee. In almost all cases the great records are the product of individuals, perhaps working together, but always within a clearly defined framework. Their names are on the door and they are quite visible to the investing public. In reality outstanding records are made by dictators, hopefully benevolent, but nonetheless dictators.

The price earning multiple must be less than ten or the inverse of the long term corporate bond rate, whichever is the less.

Try to pick a fund manager who has a well-defined strategy, has been through these things before, and go invest and stay with him.