Imagine you found a pearl – a stock that is priced at 50 but you think is worth 100. Assuming value is normally distributed around 100, with standard deviation σ=20, you think there is an 89% probability of a return of 50% or higher. With more uncertainty, say σ=30, the probability of an obscene return is a bit lower, but still high at 80%.

What more do you want? Go for it, right? Not so fast. Your valuation may be completely wrong. The market may be right: the stock is a pebble and is worth 50, with a low probability of an obscene return. Or worse, the stock may be worth less than 50. Even a lot less: if the company goes bust, the stock is worth 0.

You are like Inspector Hubbard, who is confident that Margot is innocent, but still wants to make sure that he is not making a possibly big mistake. As we have seen, the Inspector’s best approach is not to prove that Margot is innocent, but to prove that she is guilty. Likewise, the investor’s best strategy is not to prove that the stock will certainly earn an obscene return, but to prove that it won’t.

The big difference between the Inspector and the investor is that, while the first can gather conclusive evidence, the second cannot really prove anything: he can only take conclusive decisions. So his best strategy becomes: if I see such and such evidence – positive or negative – I will certainly refrain from buying the stock.

Unlike the Inspector’s proof, the investor’s decision can still produce a mistake. The rejected stock may well turn out to be a missed opportunity. But a Blackstonian investor should not be overly concerned about it. A worse mistake would be failing to see evidence against the purchase. Here the investor needs to count on his discernment, experience and discipline. He must ensure that his evidential checklist is thorough and long enough, as to minimize the risk of cumulative failure. He will decide to buy the stock only after concluding that there is not enough evidence against doing so.

Compare this against the alternative strategy: if I see such and such evidence – positive or negative – I will certainly buy the stock. This is like the Inspector trying to prove Margot’s innocence. Again, unlike the Inspector, the investor can make a mistake. But in this case a wrong decision can be much more consequential: the acquired stock may turn out to be a bad investment – the worst outcome for a Blackstonian investor. Moreover, even if he concludes that there is not enough evidence to buy, the investor will continue to believe that, while not exactly a pearl, the stock is still a rather attractive investment. As a result, he may well end up buying it anyway, thus negating the very essence of the decision process he believes he is following.

A popular concept that exemplifies this, in my opinion, ill-conceived alternative strategy is what is known as the Moat: a durable competitive advantage, or economic franchise, that allows a company to earn a superior return on invested capital for an extended period of time (for more on Moats, Pat Dorsey has written an excellent introduction).

If I see a moat – says the Moat investor – I will certainly buy the stock. At any price? Of course not. I will not overpay. I will only buy at a reasonable price – at a discount to fair value, if I can. What he is really saying is that he will buy the stock even if it is not a pearl. After all, if pearls are rare, how rare are pearls *with* a moat?

But that is where we started. We said: imagine you found a pearl. If the stock is not a pearl, why buy it in the first place? Of course, I am not saying that Moat investing is wrong. What is wrong is to look for pearls *cum* moats. And since a pearl is nothing more than a stock with an ample Margin of Safety, the Intelligent Investor faces a choice: Margin of Safety or Moat. He can’t have both.

Tellingly, the word Moat does not appear in Seth Klarman’s book.