a catalogue of errors, page-11

It's an excellent discussion topic. The best response I can give, for what it's worth, to the question of whether or not to concentrate, is a resounding it depends.

Whether it's concentration, or a preference for "high quality", or anything at all - if it is driven by a copycat approach to emulating a "great", then it's probably not going to be as successful is it could be. Thinking we have found the one true ideology, is probably self defeating at many levels. When the ideology becomes popular, then it is doubly self defeating.

For a start, we are all individuals with different skills, attributes, knowledge, character traits etc etc. It's hard enough to do well in this game being yourself, let alone trying to be someone else.

Was Lynch a better investor than Buffett? Who knows. Does the question even make sense? They were both exceptional investors. That's all that matters, isn't it?

Buffett was sometimes concentrated to an extreme degree. Did that mean his results were influenced by luck more so than for Lynch? That's a silly question. Mathematically, the question of skill can be addressed by looking at the aggregate result over many investment decisions, where no one decision was overly influential. Buffett was highly concentrated at certain points in his life, yes, but his investment results over the passing decades were the outcome of multiple decisions made over the decade. In this sense, both Buffett and Lynch demonstrated skill. The question of concentration, per se, is irrelevant.

Why did Ackman and Einhorn blow up? I don't know. Maybe they were extremely unlucky. Or maybe past success went to their head.

Mathematically, in terms of risk reduction (ie reducing the aggregate uncertainty in a portfolio) there is no value in diversifying beyond about 15 or 20 stocks (assuming approximately equal weighting are applied to each stock). As such, it is mathematical that all else being equal, a portfolio of 20 well selected stocks will perform better than a portfolio of 100 moderately well selected stocks, and that the more concentrated portfolio will carry essentially no more risk. On top of that, the more diversified portfolio will be more difficult to manage (more effort will be required etc).

But all else is almost never equal. If you are highly diversified because you keep finding outstanding investment opportunities - then why the heck wouldn't you be highly diversified? But this has nothing to do with risk reduction. It has everything to do with opportunity maximisation. This approach would require one to be highly energetic, and to have the skill and years of experience to be able to make rapid judgements with limited information (making 100% of the decision with 70% of the information). This can be highly profitable because it means you can get in quickly before the opportunity is more widely discovered. It also means of course you run a higher risk of a "bad call". But this is controlled by the allocation.

But the success of such an approach is all about the skill of the individual - and nothing to do with diversification, per se. It requires opportunities where the upside is so compelling, that it compensates for the blow ups - in the aggregate (ie attractive expected outcome). The possible outcomes here are so binary (either poor or very poor return, or spectacular success), that having a detailed understanding of the opportunity is often not required, and in fact may be counter productive.

For instance, I believe ACR is such an opportunity today. So I have invested a small proportion of my portfolio to it. You cannot remove the question of individual allocation from the question of diversification. My assessment, based on my own knowledge (which may be wrong), is that I should limit my exposure to ACR to a small proportion of my portfolio. If I were to find many similar opportunities (which I suspect I am not capable of), then I would by definition by highly diversified.

On the other hand, I believe XXX offers a high probability of an attractive return at current prices, with a very minimal probability of a poor outcome (and even lower probability of a disastrous outcome). I see no sensible reason to limit my exposure here to less than 10% of my portfolio. If I were to find a small number of similar opportunities today, then by definition my portfolio would be highly concentrated.

In conclusion, my first question was one of risk control for each individual opportunity. The degree of diversification flows from that, I believe.

There is more than one way to skin a cat. We each need to find our own way.