In this post, I will try to put together two interesting studies on portfolio construction and long term investing.

**Study 1:** Suppose I entirely focus on picking stocks that double in 3 years, how many such stocks will I require in my 10 stock portfolio to achieve portfolio IRR of 20% (Buffet standard) assuming I begin with equal allocation.

Below table gives the answer

**Conclusion**: It is ok to have a 50% error rate over a 20 year period while trying to pick “2x in 3 years” stocks. Still, one ends up with 20% IRR. **Investing is a field that tolerates very high error rates if the horizon is long enough!**

**Study 2**: What is the ideal size of a portfolio that allows a good error rate in picking stocks?

To answer this question, lets run the above table for 10, 15, 20, 25 stock portfolio sizes and compute how many “2x in 3 years” stocks are required to achieve portfolio IRR of 20% over various time frames. From the computed number, let’s compute the allowable error rate %. For example, in a 10 stock portfolio, if 9 stocks are required with “2x in 3 years” characteristics, then the tolerable error rate % is 10%, Below table shows the tolerable error rate for various portfolio sizes

Voila! there is hardly any variation in the tolerable error rate when the portfolio size is changed.

**Conclusion:** Do not overburden yourself by putting more number of stocks in the portfolio. Just try to minimize the error rate as per the above table. In fact, the more number of stocks you try to put in a portfolio, the higher chances of making mistakes and achieving poor hit rates of “2x in 3 years”. **Over a 15-20 year time frame, if 50% of your portfolio is “2x in 3 years” stocks, you will hit 20% IRR over entire 15-20 year period.**

Just buy high quality businesses at good valuations and sit tight. This is a statement seen at many places. Above explanation is numbers version of the same story. **Only ask two questions while evaluating a stock. Is it a 2x in 3 years story or 10x in 10 years story. If the answer to any of them is yes, go ahead and buy at a reasonable valuation. Else, find another** **one.**

PS: I have recently started to write a blog to keep journaling my learnings and analysis. It’s here.

You may not find anything substantial until few months, but my endeavour is to keep it up and running.