Rutherford-Darwin Portfolio – Restructuring the Management Plan Part 1: Outline of New Structure and Strategies to be Used

Tower Bridge, London

Over the years we have described a number of systems for portfolio construction and management that might be considered by anyone interested in spending a little time to manage their own money. These vary from very simple “Buy-and-Hold” portfolios, with fixed allocations – where risk is impacted by the level of diversity and correlation between the assets held in the portfolio – to more actively managed portfolios where we might be able to improve performance through adjustments made to account for changing market conditions. No simple system will work under all market conditions, and we would like to limit risk when markets do not behave in a desired manner.

In this series of posts I will review some of the ideas and strategies that we have used in the past and that I believe offer some of the best opportunities for consistent good performance in the future – maybe not with maximum returns, but in terms of managing risk and lowering draw-downs through controlling volatility. Depending on our age, investment time horizon and appetite for risk we can construct and manage our portfolios to satisfy our personal requirements without the fear of losing the wealth that we have worked, for so long, to build.

At the highest level there are basically two approaches to portfolio construction and management:

1. Trend following and momentum follow the well-known investing phrase “the trend is your friend – at least to the end”, at which point the second approach becomes more effective:
2. Mean Reversion is a contrarian approach to investing that relies on our ability to determine whether an asset might be overbought/overvalued or oversold/undervalued with price more likely to reverse direction and return to the “mean”.

On this site we have shown live examples (not cherry-picked back-tests) of both approaches – neither of which works perfectly under all market conditions.

It is difficult (if not impossible) to combine both approaches into a simple single system but my plan for the Rutherford-Darwin portfolio going forward is to try to combine them in a way that might allow us to migrate from one system to the other in a relatively smooth manner.

At the outset, my plan is to stick with a portfolio of diverse assets that include Exchange Traded Funds (ETFs) representing all the major asset classes. I have used this approach in many portfolios over the years and see no need to change it significantly since it is flexible enough to be used for “Buy-and Hold” portfolios, with fixed allocations (as used in the prior Darwin portfolio), to more actively managed portfolios (that I will be adopting going forward) where the assets chosen from the “quiver” might be in rotational favor.

At the present time I am favoring the following nine ETFs to represent the various asset classes:

1. SPLG – SPDR S&P 500 Large Cap ETF – representing the US equity markets
   • I have chosen this ETF, rather than SPY, VTI, VOO etc primarily based on it’s ~$70 price that allows me to buy more shares than for a $300-$600 Fund. I would like to hold at least 100 shares in each fund (for reasons that I will explain in future posts) and this should allow me to do that.
2. EFA – representing Developed Market equities
   • At some point I might change back to VEA, that I have favoured in the past, due to  lower expense ratios
3. EEM – representing Emerging Market equities
   • At some point I might change back to VWO, that I have favoured in the past, due to  lower expense ratios
4. VNQ – representing US Real Estate
5. TMF – representing Long-Term Bonds/Treasuries (3x Leveraged)
6. SLV – representing Commodity Metals (Silver)
   • At some point I might switch back to Gold (GLD) – this is the classic “safe haven” in metals but, historically, SLV and GLD have been highly correlated. I chose SLV because of it’s exposure to commercial applications but in the recent tariff wars this correlation seems to have been disrupted – I’ll see what happens from here, but probably no significant impact either way in the long term.
7. USO – representing Commodities/Energy (Oil)
8. IBIT – representing the Cryptocurrency markets
9. SVXY – Inverse Volatility ETF representing Volatility as an asset class
   • Normally, has a strong relationship/correlation with US Equities being based on VIX (S&P 500 Volatility) Futures prices.

I will also choose to start the portfolio with a discretionary selection of ETFs provided that they are trending upwards with positive momentum. I will cap the maximum allocation of funds in this portion of the portfolio to 90% of the total portfolio value rather than the 10% that was allocated to the initial “Buy-and Hold” portion of the previous Darwin portfolio. The additional risk will be managed through Risk Parity and volatility targeting. This will be explained in future posts as the portfolio builds.

I will start by going back to the Kipling Workbook that we developed a number of years ago and focus on the worksheet that we developed for the BHS (Buy-Hold-Sell) strategy/model. After the Close on Thursday (15 May) I took a look at the recommendations from the BHS momentum strategy/model built into that workbook:

The BHS model incorporates measurements of “momentum”, over long- and short-term time frames and weights them in an algorithm to rank and recommend ETFs with the highest “Scores”.

From the above screenshot we see that Developed Market Equities (EFA), Emerging Market Equities (EEM), US Real Estate (VNQ) and the Cryptocurrency ETF (IBIT) were suggested Buy candidates.

There is a lot of information in the above screenshot, that evaluates the parameters used in the model, including longer term momentum as measured by the slope of the linear regression line going through the daily closing price points, over 60-Day and 100-Day look-back Periods (Columns LRS1 and LRS2 respectively) and shorter term (5-Day and 8-Day) momentum as reflected through the sign and magnitude of Heikin-Ashi Candles (Columns HA5 and HA8). All these parameters are “measured” and weighted to generate a “Score” and “Rank” for each ETF

The following screenshot shows a visual picture of rolling 5-Day and 8-Day Heilin-Ashi candles:

As we can see, VNQ has been in a bullish up-move over the past ~3 weeks.

At one time we used this model extensively to manage a number of our portfolios here at ITA Wealth but, with the wide acceptance and “popularity” of momentum as a strategy, there is a suggestion that, whilst still a valid model for investing, that a little of the “edge” has gone out of the strategy.

It is therefore prudent to look a little more closely for confirmation of the recommendations identified in the worksheet, such as the position of current price to a moving average of price, and/or the relationship of one moving average to another, on a different time-scale, to provide other possible methods of identifying the current “trend” of an asset. This kind of information is seen in tabular format in the top screenshot as related to the 13-, 49- and 195-period Exponential Moving Averages (EMA).

Using different software to plot price charts we might see something like the screenshot shown below to add to our analysis/evaluation and to provide additional confirmatory signals:

In this daily chart we see current price closing above both the 8-Day and 21-Day moving averages with the short-term moving average (blue line) above the longer-term moving average (red line). Both conditions suggest/confirm that VNQ (in this example) is in a short-term (~1 month) bullish uptrend. However, we also see a longer-term downtrend (shaded light green channel – top left to bottom right). At some point we have to make a decision as to whether the trend that we are interested in following has changed. In the above example we might decide that we have changed from a downtrend to an uptrend since price has closed outside the upper boundary of the longer term downtrend channel in addition to being in an uptrend as defined by price and shorter term moving averages. But, of course, we have to recognize that we might be wrong and price might still turn around and fall back into a downward trend – and we need an action plan as to what we might do should it do so. My choice of using the 8-Day and 21-Day moving averages in the above chart is arbitrary in that it simply reflects an intermediate term measurement period that seems appropriate for the frequency at which I might be comfortable adjusting my positions in the portfolio – and because I have an obsession with Fibonacci numbers 😊. However, I could just as easily have chosen the 13-Day (also a Fibonnacci number) and 49-Day lookback periods used in the Kipling workbook. These numbers should not be optimized through back-testing other than to confirm that performance is not strongly dependent on which numbers are chosen – i.e. we need to look for a “robust” system that we can stick with.

Other indicators also provide information on trend and strength/momentum and two of these indicators, the Moving Average Convergence/Divergence (MACD) indicator and the Relative Strength Index Indicator (RSI) are plotted in the lower panels of the above screenshot. In the price plot the candles are colored dark blue or red when these indicators are in agreement and associated with a bullish (blue) or bearish (red) trend. Paler (blue or red) candles indicate that there is a conflict in the signals being generated by these two indicators and that they are not in agreement.

We might also look at the volume of shares traded as a possible indication of institutional buying or selling on high volume above the daily average.

Based on these tools for analysis and confirmation I have chosen to open positions in all four of the ETFs recommended as Buys in the Kipling workbook.

On Friday I purchased 200 shares of EFA, 275 shares of EEM, 180 shares of VNQ and 130 shares of IBIT:

Having outlined how I am constructing the portfolio I will follow up with an outline/description of how I plan to allocate funds, and to manage risk, in future posts.

Checking on the workbook recommendations after the Close on Friday, SPLG triggered as a recommended Buy so, with all the confirmatory signals/indicators showing favorably, 100 shares of SPLG were purchased for the portfolio this morming (Monday 19 May).

Summary:

In addition to the level of funds allocated to the strategy (maximum 90% allocation rather than 10%) the other significant change here is to revert back to a more actively managed strategy rather than adopting the “Buy-and-Hold” approach used for the initial portfolio – although the  assets in the quiver, for possible inclusion in the portfolio, will remain the same.

Readers requiring additional explanations of the rules/conditions described above are encouraged to ask for clarification in the comment section below.