Building a Better Portfolio

Introduction

In this blog post we are going to discuss how to build a better portfolio using two key Stock Rover tools: Portfolio Analytics and Portfolio Trade Planning.

Note: none of the stocks mentioned in this blog post constitute a recommendation to buy or sell. You should always do your own research before committing your hard earned capital to any security.

Also Note: Portfolio Analytics and Portfolio Trade Planning are available with the Premium and the Premium Plus Stock Rover subscription plans.

Rebalancing and Trade Planning

One of the most powerful and probably underappreciated features of Stock Rover is the Portfolio Trade Planning and Rebalancing facility. This facility actually provides two distinct capabilities: Trade Planning and Rebalancing, which address portfolio planning and modeling from different perspectives.

Specifically, Trade Planning is building a model portfolio that shows the desired end result, position-wise, and the set of trades necessary to achieve that end result from an existing portfolio. Trade Planning is generally used for portfolios where there are no specific allocation rules, but rather you buy and sell individual securities with the goal of maximizing portfolio performance, given the level of risk that portfolio is willing to take.

Rebalancing is when the model portfolio has a certain well-defined allocation of positions (stocks, ETFs and/or funds) by percentage. Rebalancing shows how far the actual portfolio has drifted from the ideal balanced portfolio and the trades necessary to bring the portfolio back in balance.

For the purposes of this blog we will be focusing on the Trade Planning tool. A separate blog post discusses Portfolio Rebalancing.

The Trade Planning tool in Stock Rover is found by selecting Rebalancing as shown in the screenshot below, as the selected tool can perform both the Rebalancing and Trade Planning functions.

Improving Our Portfolio

We will begin by working with an existing portfolio and using the Trade Planning tool along with Portfolio Analytics to help us plan out a new set of trades that will ideally improve the performance and risk characteristics of the portfolio.

Our example portfolio will be a value portfolio I created on June 18th, 2019. The portfolio contains 21 stocks, each staked to a $100,000 initial investment. Since the portfolio has, as of this writing, been running for almost 6 months, it is a good time to take a look and see how it is doing and whether any changes may be in order.

I’ll start by analyzing the portfolio performance as a whole over the last 6 months using the Stock Rover Portfolio Analytics tool. The first screenshot assesses overall portfolio performance since inception.

We can see that the portfolio has performed well, returning 9.5% in its initial 6 month period, vs. 7.1% for the S&P 500. However, to achieve that performance, it took on more risk than the S&P 500, as the portfolio has a one year volatility of 0.20, whereas the S&P 500 is at 0.14. Corroborating that increased risk is the Beta, which is 1.26, which means the portfolio is going to move more than the S&P 500 on an average day. These characteristics help juice returns in an up market, but will not be so wonderful in a down market. Indeed the risk-adjusted return vs. the S&P 500 is actually -0.6%. This level risk may not be what you are after in a value portfolio.

Position Performance Analysis

Let’s take a look at how each of the 21 positions performed.

Here we can see there has been a wide variety of performance within the 21 stocks that constitute the value portfolio. Looking at the screenshot above, with the table sorted by percentage of total return, we can see that four of the 21 stocks provided the lion’s share of the overall portfolio return. Specifically, the top four stocks provided 2/3rds of the overall return of the portfolio. Micron (MU) was by far the portfolio MVP by providing 28.9% of the portfolio’s return on its own.

On the other side of the ledger, five of the stocks provided negative return to the portfolio’s overall performance.

When reviewing a portfolio with the idea of making changes, you should look at all of the stocks, but generally the outliers are the best place to start. The best performers because they may have come too far too fast and perhaps momentum will end soon. And the worst performers because perhaps they are just dogs and we are better off finding stronger stocks. With this in mind, let’s look at some more data.

First let’s look at the short and long term performance history of each stock in the portfolio vs. the de facto benchmark, the S&P 500. The screenshot below shows this data. I have sorted the table based on return vs. the S&P 500 over the last 6 months, as that period is close to the period from portfolio inception to now.

Note that on the dog side, the bottom few stocks have a lot of red, indicating that they never beat the S&P 500 for any period (Lincoln Financial, Molson Coors Brewing) or almost never beat it (Norwegian Cruise Line, Lear, Prudential Financial); these are candidates to trade out of, under the theory that if you are a dog that long, then you are just a dog.

On the other hand, companies such as Steel Dynamics, Morgan Stanley and Capital One Financial also have been poor long-term performers vs. the S&P 500, but at least are now showing strength, the ideal case for a value stock, as perhaps they are being discovered. These would be candidates to hold.

For the top performers, it looks like Owens-Corning, Lennar and Toll Brothers may be losing momentum. They would be candidates to reduce or sell, just based on their loss of momentum, as judged by their historically strong, but recently weak, performance vs. the S&P 500.

Deeper Research

So now that we have these potential candidates to trade, the idea would be to do deeper fundamental research on each of these companies to see whether or our original performance and momentum based concerns correlate with deeper concerns regarding the outlook and the expected future financial performance of the company. The outlook for the company’s industry must also be taken into account, as it is difficult to swim upstream against the flow.

The actual research required is beyond the scope of this article. But the good news, is we have developed a mini course to help you do fundamental research with Stock Rover.

Trade Planning – Selling

Let’s assume that we have done our financial research and we have decided to sell all the poor performers and half of the allocation of our winners that we have concerns with. Plugging all of this into the Stock Rover Trade Planning tool nets the following trade plan.

Notice the “Plan By %” box is unchecked. This allows us to work with the share counts rather than percentages. If the box is checked, you plan by specifying desired percentages rather than desired quantities of shares and Stock Rover computes the shares to trade.

Wow, that’s a lot of sell. Now we need to find some buy candidates. Before we do that, let’s take a look at our sector allocations to get a feel for sectors we may want to add to ensure proper sector diversification. This is done within the Trade Planning facility as shown below.

We can see that the “Planned Portfolio”, which is the current portfolio incorporating our planned sales, is actually reasonably balanced. Perhaps a bit heavy in financial and consumer cyclical and a bit light in technology and healthcare, but not bad.

Note there are no energy stocks, utilities or industrials. Energy stocks have been terrible performers for a long time. Utilities have appreciated so much that they have generally moved out of what would be considered value territory. The oversight in industrials was probably a portfolio construction error on my part. We have a small allocation of consumer defensive, but similar to utilities, consumer defensive stocks have generally left the realm of value.

So when we go shopping for new value stocks, we will be open to any sector, but will be biased towards adding industrials, technology and health care.

Trade Planning – Buying

Let’s assume we have run our value screeners, done our fundamental research and have come up with eight new names we want to add to the portfolio. They are as follows:

Ticker	Company	Sector
ABBV	AbbVie	Healthcare
CNC	Centene	Healthcare
CVS	CVS Health	Healthcare
ETN	Eaton	Industrials
HUBB	Hubbell	Technology
PEP	PepsiCo	Consumer Defensive
UTX	United Technologies	Industrials
WRK	WestRock	Consumer Cyclical

Unfortunately, our selling spree will not net us enough money to add eight new value positions at $100,000 each, so we will use a slightly under 100K equal allocation to all of the new positions. Then we can do analytics on the full set of trades (buys and sells) to see what the impact on the portfolio is regarding portfolio price characteristics, fundamental metrics and sector allocation.

The Trade Plan

First let’s look at our new trade plan:

We can see that there are a lot of trades. We are doing extensive remodeling to our value portfolio. The second to last columns, Planned Trades shows the quantities to buy and sell in order to realize our shiny new value portfolio.

A couple of practical tips on the actual buying and selling. I usually intermix selling and buying trades, always selling enough first to ensure there is enough cash so that subsequent buys are covered. So I might sell two, buy one, then sell, buy, sell, buy until the plan is complete. Also, I often have to adjust the quantity of the last buy to ensure my final cash balance remains positive, as prices do shift during the trading day.

Analytics

OK, so now that we have built a plan, how does it look? Let’s start with the Analytics, which show key price performance characteristics of the current portfolio and the new trade planned portfolio.

Here we see some useful information. Using the period that the portfolio has been in existence, we can measure the characteristics of the actual portfolio vs. how the new portfolio would have performed. We see some good things. The historical return would improve by 2.1%. Almost as good, the maximum drawdown also improved from a worst case loss of 12.6% to 9.7%. So better performance with less risk; this looks very promising.

Exploring further, as expected, our risk-adjusted return vs. the S&P 500 has improved, and our beta and volatility have both decreased slightly. And our Sharpe ratio, another measure of risk-adjusted return has improved considerably. I like the look of these changes so far.

Fundamentals

Ok, the analytics look great. How about the composite fundamentals of the portfolio. Here we use the Metrics tab of the Trade Planning tool.

Here we can see that the decisions we made have made the portfolio a little less “valuey”. The composite P/E ratio of the portfolio has increased from 11.3 to 13.5. Still well below the market multiple, but not quite as cheap.

On the good, we have substantially increased the expected earnings growth of the portfolio from -10.4% to -1.6%. However we have decreased next year’s expected earnings growth of the portfolio from 15.3% to 13.2%. Next year’s numbers are still close, and there is always a lot of uncertainty and change in next year’s projections, so this minor decrease is not troubling.

Price to Sales has stayed the same at 0.7, but like P/E, Price to Book has also increased, from 1.2 to 1.6. This proposed portfolio is definitely moving a bit away from the value stable relative to the original portfolio. To the good, our dividend yield has increased from 2.1% to 2.3%.

Sectors

Finally let’s take a looks at the sector allocations of the proposed portfolio vs. the current portfolio.

We can see that the current portfolio is quite heavy in financials and consumer cyclical and light in tech and healthcare with a non-existent representation of industrials. The planned portfolio aims to fix this, and fix this it does.

We can see much better sector balance in the new portfolio. The only area one may quibble with is that we are a bit heavy in healthcare, but this sector has performed poorly this year and is inexpensive relative to its normal valuation vs. other sectors. So perhaps for a value portfolio, this may be a good bet to make. Still the sector bet is 22.2%, which isn’t huge, whereas the current portfolio has two sectors that are almost 60% of the overall portfolio, with financial alone at almost 32% of the portfolio. So the planned portfolio is definitely much better in this regard.

The portfolio allocation tab shows this graphically with a pie chart.

So we are done. All that is left is to follow the trade plan and do the trades at the brokerage.

Final Thoughts

I very much like this new value portfolio vs. the original value portfolio I planned out 6 months ago. The Trade Planning tool has helped me construct the new portfolio, validated that the changes will be beneficial in a number of ways (performance, risk, sector allocation) and also indicated a few ways the portfolio has strayed slightly from its value roots, which I can live with, as the portfolio is still solidly value.

In the real world I would continue to tweak this portfolio a bit to see if I could improve its value metrics without harming its much improved risk reward profile and sector allocation.

In any case, once I am done with Trade Planning and am satisfied with the new model portfolio, the actual trading at the brokerage becomes dead simple. Just follow the trading recipe. All emotions associated with executing the trades are gone, which, when trading, is a very good thing indeed.