Our Model Portfolio is Outperforming the S&P 500

Our investment approach has produced much better returns than the benchmark S&P 500 this year.  This is due to two components: 

• Our model portfolio is outperforming.
• We advised buying triple leveraged ETFs when the market was near its lows in March (this will be covered in a separate blog post).

Earlier this year we published a model portfolio of 40 stocks and ETFs with the stated goal of outperforming the S&P 500 index.  This analysis was done in mid-February, before the stock market began to drop due to the coronavirus.

We recently calculated the performance of this model portfolio, and are happy to report that it has outperformed the S&P 500 over the past five months.  We first published the portfolio on February 19^th after the close.  Therefore we have calculated performance from 2/19/20 to the most recent month end, which was 7/31/20. 

The portfolio declined by -1.2% during this period, while the S&P 500 declined -3.4%, meaning that the portfolio had an excess return of 2.2% in just over five months. 

This performance was achieved with absolutely no turnover, holding the same group of stocks as the market went through the large V-shaped move that has occurred this year.  Obviously this year has been a very unusual market environment, and most money managers are likely to have made some changes in portfolios since February. 

In fact we identified a number of valuable ideas for buying and selling at various times over the past five months, such as when the market formed a “follow through” day on April 6^th.  However because we did not publish these trades on the blog, we will not include them in the discussion of the model portfolio.                    

In designing the portfolio in February, our analysis included several elements, including the following: 

• Equal weight the largest stocks in the S&P 500: A handful of giant capitalization stocks such as Amazon (AMZN), Apple (AAPL) and Microsoft (MSFT) account for roughly 20% of the total weight of the S&P 500.  Therefore to eliminate the risk of underperforming if these stocks did well, we included these stocks in the model portfolio with weights equal to their weights in the S&P 500.

• Sector weights: Relative to the S&P 500, we overweighted the technology sector, and in order to do this, slightly underweighted several other sectors, namely industrials, consumer cyclicals, consumer staples and financials.  This sector analysis was based on a variety of considerations, including trends in relative strength.

• Industry groups: We overweighted the portfolio in selected industry groups, including residential construction and software.  These groups were chosen because numerous stocks in each group had very attractive characteristics. 

• Stock selection: Our stock selection process is based on a combination of fundamentals and technicals.
  • For the fundamental data, we are using a quantitative stock selection model with a proven track record, based on factors such as earnings growth and earnings estimate revisions.
  • For technical analysis, we are using state-of-the-art charting software to scan the market and identify stocks with bullish technical characteristics.    

• Risk is low because of diversification: Despite the overweightings in certain industry groups, the portfolio is highly diversified, containing 36 stocks and four sector ETFs.

Please see the spreadsheet below for more detail.  The best performing stocks in the portfolio have been Amazon (AMZN), Apple (AAPL), Activision Blizzard (ATVI), Synopsys (SNPS) and Fortinet (FTNT).  The worst performing were OneMain (OMF), the Energy Sector ETF (XLE), Intel (INTC), Caci International (CACI) and Sun Life Financial (SLF). 

This data highlights the importance of sector and industry group weights.  The technology sector has been the strongest sector in the market over the past five months, and we were overweighted in technology.  By comparison, the financial sector has been very weak, and we were underweighted in financials.

Figure 1     Performance of Model Portfolio — First Published in February 2020