That's not "a portfolio of the 20 or 30 most profitable" companies vs the S&P 500. That's a portfolio of ten companies that were at one time the most profitable and then not regularly adjusted compared to an index that was updated regularly (only 312 companies in the index in 2005 were still in it in 2015[1]).
A realistic portfolio to prove/disprove your theory that investing in highly profitable companies is a losing strategy would be to take the 25 most profitable companies (I said 20 or 30 so let's just go with the middle) and update it every year.
With only the eight most profitable making up ~15% of the weight a portfolio of 25 is probably going to approach 30%. It's hard to imagine the largest components representing nearly a third of the index weight are going to move completely and drastically divergent to the index as a whole.
A realistic portfolio to prove/disprove your theory that investing in highly profitable companies is a losing strategy would be to take the 25 most profitable companies (I said 20 or 30 so let's just go with the middle) and update it every year.
With only the eight most profitable making up ~15% of the weight a portfolio of 25 is probably going to approach 30%. It's hard to imagine the largest components representing nearly a third of the index weight are going to move completely and drastically divergent to the index as a whole.
[1] http://marketcapitalizations.com/changes-in-sp-500-component...