Investing builds upon trust. And the same way we like to know our friends really well before borrowing them money, we should understand our investments before committing money to them. Throughout the site, we characterize our portfolios with the help of charts and metrics. In this post, we explain which of these we use and how to interpret them.
The primary objective of investing is to make a profit. Consequently, the return on investment is the first thing most investors wonder about.
In its purest form, we invest a dollar amount, wait patiently for our investment to grow, and get paid back a larger amount. Here is how this looks like in our metrics:
The example shows how our Pick Me Up strategy grows an investment of $1,000 made in January 2007 to more than $4,500 in April 2020. To put this number in context, we use benchmarks. In this example, we use SPDR’s S&P 500 ETF (SPY) as a reference, learning that Pick Me Up outperformed the benchmark.
While these absolute numbers are very intuitive, they are not always easy to compare. To address this issue, we calculate the Compound Annual Growth Rate, often abbreviated as CAGR.
This number expresses the growth of our investment as an annual percentage. Over the simulation period, our Pick Me Up strategy grows by a little more than 12% per year. This number assumes that all gains stay invested; they are continuously compounded.
We can find the same information on the Equity Chart. Horizontally, we see the time axis: starting in January 2007 and ending in April 2020. Vertically, we have the dollar value of our investment: starting at $1,000 and ending at $4,500.
Please note that we scale the vertical axis logarithmically. This type of scaling has several advantages over linear scaling:
- Equal growth factors, e.g., a doubling in value, result in the same vertical distance: the distance from $1,000 to $2,000 is the same as the distance from $2,000 to $4,000.
- If an investment grows at a constant rate, the resulting chart shows a straight line.
- If two investments grow at the same rate, their equity curves are parallel.
The CAGR cannot be found directly on the chart. However, we find it represented by the slope of the equity curve: the steeper the slope, the higher the growth rate.
Most investments fail to provide steady returns with a constant growth rate. Instead, their growth rate fluctuates, making it difficult to read from the chart. Here, a bar chart showing the annual returns comes handy. Horizontally, we see the calendar years, and vertically we plot the growth rate for the respective calendar year.
While this is a familiar and handy representation, it has some shortcomings. For once, it does not show what happened between the beginning and the end of the year: this path could have been all but smooth. Also, any gains or losses close to the end of the year induce some misleading jitters here.
Losing money is a painful experience. Therefore, many of the charts and metrics aim to measure investment risks.
As we have seen above, returns are not constant but fluctuate over time. A measure for this fluctuation is the Standard Deviation of Returns or Volatility.
This number helps us estimate a range of likely investment returns. Our Pick Me Up strategy has a Compound Annual Growth Rate of about 12%, and its Standard Deviation of Returns is almost 16%. Statistically speaking, we have a likelihood of 68%, that Pick Me Up will return between -4% (12% minus 16%) and +28% (12% plus 16%) annually. While there are no guarantees, the bar chart above confirms that our Pick Me Up returns stayu close to these bounds.
We like to point out that for many strategies, the Standard Deviation of Returns is higher than the Compound Annual Growth Rate. In other words, volatility drowns trend, and there is a substantial risk of losing money over a given 1-year period.
We can also find the same measure on our Equity Chart. If CAGR is the slope, then the Standard Deviation of Returns is the width of a corridor, within which the equity curve ranges for most of the time.
From a mathematical point of view, these may be useful measures, but many investors think differently: Investments are supposed to grow, and any reversal from that upward trajectory is considered a loss.
Underwater Charts reflect this view: they chart the percentage lost since a previous all-time high over time, the so-called drawdown. From the chart above, we learn that our Pick Me Up portfolio suffered its worst loss in late 2008, which was about 25% from its previous high. We can further see that the longest drawdown started in October 2007, and lasted until January 2010.
Our metrics show this information as follows:
In addition to the Maximum Drawdown, we also measure the Maximum Flat Days. This number is the longest time it took from one peak to making a new all-time high. With 805 days, this is more than two years. And even though that is a long time, this is still less than half of the benchmark.
We calculate Maximum Drawdown at the end of each day. This leads to worse results than calculating drawdowns only at the end of the month.
Because the index uses squared drawdowns, it applies larger weights to more profound losses. We can interpret the Ulcer Index as a representative drawdown, which our investment will reach frequently.
The chart above shows the Ulcer Index for our Pick Me Up strategy. We see that the portfolio reaches or exceeds this drawdown level regularly.
When working a regular job, we get paid for performing our duty. Similarly, as investors, we get paid for taking on risk, the so-called risk premium. By dividing a return measure by a risk measure, we calculate Risk-Adjusted Returns: measures expressing how well we are compensating for taking on risk.
The most commonly used measure for risk-adjusted return is the Sharpe Ratio. At its core, it divides the Compound Annual Growth Rate by the Standard Deviation of Returns. However, there is an additional twist: before performing these calculations, we subtract the Risk-Free Rate of Return from the profits. Doing so makes sense, as we are looking for a measure of taking on risk, and receiving risk-free interest is not that. In today’s environment of low interest rates, the risk-free rate is close to zero. However, this has not been the case in prior decades.
There are many ways to calculate the Sharpe Ratio, which make this measure hard to compare. For TuringTrader.com, we chose to use monthly returns, and annualize the results. For the risk-free rate, we use the Secondary Market Rate for the 3-Month Treasury Bill.
Another important measure for risk-adjusted return is the Ulcer Performance Index or Martin Ratio. This measure divides the CAGR by the Ulcer Index.
Our example shows how, according to the Sharpe Ratio, the Pick Me Up portfolio produces more than twice the return per unit of risk as a straight investment in the S&P 500 would. According to the Ulcer Performance Index, our Pick Me Up strategy achieves almost four times the risk-adjusted return of the S&P 500. When comparing portfolios, it is good practice to watch both measures.
Investors often believe that to achieve high returns, they have to take on high levels of risk. Risk-adjusted returns debunk that myth. Investments with similar risk profiles may lead to very different profits, and Tactical Asset Allocation augments this effect.
Mathematically speaking, Beta is a measure of correlation. It is part of the Capital Asset Pricing Model and describes how much an investment moves in tandem with the market. In bullish markets, Beta may seem desirable, as stocks with a Beta larger than one will provide amplified growth. In the context of portfolios, Beta is typically seen as a risk factor, though.
This makes sense, as a well-rounded portfolio should appreciate value throughout all economic seasons and independently from the movements of single asset classes. On our site, we always measure Beta against the S&P 500, as stocks are the driver of growth in most portfolios, and also the most significant contributor to portfolio risk.
Simulations of our strategies with historical data, so-called Backtests, are useful methods to analyze how the portfolios performed under a variety of past market conditions. However, even though markets go through cycles, history does not repeat itself exactly. Monte-Carlo Simulations address this issue by calculating the probabilities of achieving specific results based on the outcome of random experiments.
On the horizontal axis, our charts show cumulative probabilities. On the vertical axis, we show CAGR as well as Maximum Drawdown.
For CAGR, the charts show the probability of returns less or equal to the value on the vertical axis. The example shows that for Pick Me Up and a simulation period of 13 years, the probability of annual returns less than 6.2% per year is only about 10%. Or, with 90% confidence, we can expect profits to exceed 6.2% per year. In comparison, an investment in an S&P 500 ETF may yield negative returns over the same period with a probability of 10%. Strategies with higher average growth rates sit higher in these charts. Equally important, less sloped charts indicate more predictable returns.
For Maximum Drawdown, the charts show the probability of maximum losses less or equal to the value on the vertical axis. With a probability of 90%, our Pick Me Up strategy has maximum drawdowns of less than 39%. In comparison, the S&P 500 ETF has a much higher risk: with a likelihood of 10%, the maximum drawdown will exceed 63%. Again, strategies flatter slopes indicate more predictable results. Further, strategies with a shallower profile have an overall lower risk.
Portfolio Turnover is an important measure: for once, it is closely linked to the taxation of capital gains. Further, frequent turnover implies trading costs, both in the form of fees or commissions, and in terms of inefficiencies through slippage or bid/ ask spreads.
Our Exposure Chart shows how the strategy allocates capital to its assets. From the above chart for the Pick Me Up portfolio, we can see the following:
- The strategy continually invests 100% of the available capital.
- Most of the time, the strategy invests about 85% in an S&P 500 ETF and the remainder in a leveraged S&P 500 ETF. The majority of the S&P 500 exposure is held for multiple years.
- In times of market stress, the strategy gradually reduces exposure to the stock market and invests in an ETF of long-term treasuries.
These charts give us a very detailed view of how long the strategy holds its positions and what the size of these positions is. Unfortunately, the charts become confusing for strategies using more than three assets.
Therefore, we also use charts visualizing the holding period of positions. The chart above shows that the strategy holds many positions for 365 days and longer. Unfortunately, these charts do not provide any indication of the position size and therefore become inconclusive for portfolios that continuously resize their positions.
Benchmarks help to analyze strategies by putting the results in context. There is no set rule which benchmark to use. Generally, we use benchmarks with a similar level of risk as our investment, but not necessarily the same asset classes.
For aggressive stock growth strategies, we typically use a proxy for the overall stock market. Specifically, we use S&P 500 ETFs, which capture capital gains as well as dividends paid. This benchmark reflects that typical growth portfolios build on stocks, and is relevant because we expect the strategy to have a positive correlation to the stock market. Further, the benchmark allows a direct comparison of the strategy’s ability to handle recession periods.
For less aggressive or cross-asset strategies, we use a Vanilla 60/40 portfolio. This benchmark reflects the goal of reducing portfolio volatility and is relevant because stock/bond portfolios are the ubiquitous approach to portfolio construction.
TuringTrader.com offers a steadily growing selection of portfolios and strategies, along with background information, references, and daily updated charts and metrics. We very much hope that this guide helps you digest this wealth of information.