In the school of money management, there are a number of different approaches. Most believe managing a portfolio consists of some percentage art and some percentage science.
I suppose the science part could be relegated to a computer model. However, the portion of portfolio management which is dedicated to "art" requires a human being. Off the top of my head, I can think of at least one ETF and one mutual fund that discounted the "art" portion and relied solely on a computer model to make all decisions. In these cases, I learned a valuable lesson: you cannot and should not discount the human element. In each case, there were significant periods of underperformance. Hence, I believe that a good quality money management approach requires some degree of art and science. Then comes diversification.
Can a portfolio be too diversified? I used to think not. But today, I believe the answer is yes. Consider this. Investment A (the S&P 500 Index) and Investment B (S&P 500 Bear Market Fund) react exactly opposite to each other. In essence, they are perfectly negatively correlated. Therefore, as A rises, B falls, and vice versa. At the end of a specific period, your portfolio essentially has a zero return. This is because they cancel each other out. To correct this, you would have to select two investments with positive returns over the exact same period. Assume C and D both have an average three-year return of, say, 8.0%. In this case, the portfolio would have a positive return, but it would be impossible for these investments to have a correlation of negative one.
A few years ago, I wrote about a subscription I had with a piece of software called GSphere. This program provided a mathematical measure of a portfolio's diversification, which is a great idea in theory. Certainly, you would want to be very diversified during a bear market. However, the more diversified you are when markets are rising, the weaker your returns will be. For example, this program would score the portfolio from 0% to 100%. A portfolio with a diversification score of 100% would be the most diversified, and of course, the safest. However, a 100% score exists mostly in the classroom. In reality, a score of 55% is considered to be very well diversified.
I still believe diversification is important. Today, I select good-quality funds with strong three- and five-year track records, and it's important that they don't move in lock step. If they did, then your diversification score would be very low. Of course, this wouldn't be a problem during a bull market.
And so the subject of portfolio management continues.