Portfolio Development Series - Part 2

Planning The Portfolio Objectives, Metrics, and Trading Systems

Welcome to the “Systematic Trading with TradeQuantiX” newsletter, your go-to resource for all things systematic trading. This publication will equip you with a complete toolkit to support your systematic trading journey, sent straight to your inbox. Remember, it’s more than just another newsletter; it’s everything you need to be a successful systematic trader.

---

Introduction:

In the first article in this portfolio development series, we discussed the concept of portfolio-level thinking. I took a simple momentum system and applied portfolio-level thinking to show how the performance of one individual trading system does not matter as much as many believe. What really matters is how that one trading system contributes to a portfolio of many trading systems.

That introductory article was written because understanding portfolio-level thinking is critical to developing a portfolio of trading systems. Without this understanding, we could potentially throw away perfectly good systems that look ugly on their own but would otherwise make a great addition to a portfolio of many trading systems.

In this article, I want to start planning out what the final systematic trading portfolio might look like. I will lay out a general plan and direction with some portfolio-level performance goals. We don’t want to confine ourselves to a design space that’s too specific, but knowing the general direction will really help in figuring out what needs to be developed to create this systematic trading portfolio. This plan and set of goals will help us develop a well-rounded equities portfolio. So, let’s dive in and start planning.

---

High Level Portfolio Design Items To Consider:

Before diving into specific trading system ideas or portfolio goals, we need to define the high-level direction the portfolio will take. We can’t just start developing a portfolio without considering the things I will discuss below.

We need some sort of direction in terms of market type/asset class, market region, timeframe, and side (long/short/both) for which this portfolio should be designed.

So, first things first, what market type/asset class will this portfolio be designed for?

Market Type/Asset Class:

• Equites

• Crypto

• Futures

• Forex

• Options or Shares

• etc.

You could create an entire portfolio for each market type/asset class listed above (and I’m sure there are others too, which I failed to list). You could also trade each market type/asset class through the lens of shares or options.

Each market type/asset class has its own pros and cons, but as equities traded with shares are my specialty, we will focus there. Trading equities with shares is also likely the most familiar to the majority of people as well.

One thing to note in case you are not interested in trading equities via shares, these portfolio development principles apply to any market type. So, everything discussed throughout this Portfolio Development series will be applicable to whatever sandbox you want to play in. You could even develop a portfolio for each market type/asset class, where each portfolio contains multiple trading systems, essentially a portfolio of portfolios concept.

Example:

• A portfolio of 5 shares trading systems on equities

• A portfolio of 3 spot trading systems on Crypto

• An options strategy portfolio on equity ETFs

• setc.

For those who are not sure which market type/asset class they want to trade, here are a couple of questions consider. Under each question I added some examples of things to think about:

How stable is the market type/asset class?

• As an example, equities exchanges are very reliable, but crypto exchanges may not be. However, there is likely more edge in crypto than equities.

• Crypto has significantly more volatility and downside risk than equities. More care needs to be taken with crypto to ensure risk is accounted for.

• With that said, with more risk there is generally more reward. The less mature or stable the market type/asset class the more alpha there will be laying around.

How well do you understand the market type/asset class and its dynamics?

• You may be very comfortable with trading equities shares and understand how they work, but futures and options are a completely different ballgame for you.

• In that case, it may make more sense to start with what you know and are familiar with, rather than spending a lot of extra time trying to learn a new market type/asset class.

• You can always transition to more sophisticated methods on more complex or less familiar markets later, but start with what makes the most sense and has the lesser barrier to entry.

How sophisticated are the market players?

• Futures, forex, and large-cap equities tend to be dominated by massive hedge funds and banks. These are sophisticated players with access to much more data, tools, and resources than you.

• Starting in a less sophisticated market like crypto, small-cap stocks, or smaller countries' stock markets may be a better option as a starting point.

• Think of this like sitting at a poker table full of professional players versus sitting at a table full of all your drunk friends. The more sophisticated the market players, the harder it will be to extract an edge.

How easily can you access the market?

• For a while, certain crypto exchanges were banned in the region I live in. This made it extremely difficult for me to trade the crypto market. I could only access exchanges with maybe 25–50 coins listed, whereas in reality there were tens of thousands of coins across the entire crypto market; I just didn’t have access to them.

• These restrictions made is so I either had a very limited trading universe or had to take massive risks to access a larger universe, such as signing up for an exchange with a VPN to a different country and risking having my funds locked If the exchange found out.

• On the other hand, I have access to several countries' equities markets, and the barrier to entry is nothing more than a few button clicks with my broker to request access to the market. I wanted to trade crypto, but it made much more sense for me to focus on equities in this case.

As outlined above, there is a bit more to it than just randomly picking a market to trade. There are practical considerations that account for ease of access, legality, the size of the edge that exists, and overall market risk. Give those points some real thought before randomly picking a market type to trade.

Next comes the specific market regions to trade. This category will be slightly different depending on what market type/asset class you choose. Since this article series is directed toward equities, we will focus there:

Equities Market Regions:

• North America

  • USA Stock Market

    • Nasdaq 100

    • Russell 1000

    • S&P 500

    • Etc.

  • Canadian Stock Market

• Australian Stock Market

• European Stock Markets

  • France

  • Germany

  • Norway

  • Sweden

  • United Kingdom

  • Etc.

• Asian Stock Markets

  • Hong Kong

  • Japan

  • Singapore

  • Etc.

• South American Stock Markets

  • Brazil

  • Etc.

• Etc.

There are so many options to choose from in terms of country stock markets available to trade. A pro tip: don’t just go to the USA stock market just because it is the biggest and most liquid. The liquidity and numerous stocks to choose from is great, but the US stock market also has the most sophisticated players.

You can extract more edge from smaller stock markets (or smaller cap stocks within the USA stock market). This is not to say don’t trade the USA stock markets—I definitely do, and there are real effects/inefficiencies to extract there—I’m just saying do not box yourself into only the USA stock market.

The next set of items to consider are the timeframes you want your trading systems to use. There are some real considerations here. For example, trading an intraday system on 5-minute bars may not make sense for people who can’t monitor the system all day. This is because if something goes wrong (e.g., internet goes out, bugs in the system, API data feeds drop, orphaned positions, etc.) you need to be in front of your computer at all times to fix the issue. Also, very short timeframes are much more susceptible to the impacts of slippage and commissions, so your fills have to be near perfect.

I have a day job, so intraday strategies do not make sense for me. Believe me, I tried, and I failed at it. My internet went out, systems crashed, by broker platform crashed etc. all while I was at work; not a great situation. Having a day job and trading intraday is extremely difficult.

I now focus on the daily timeframe and higher, as the pace of the trades is something I can handle with a day job. I can also better account for things like internet outages and orphaned trades because I have a lot of time between market close and market open the next day to reconcile any issues.

Timeframes:

• Intraday

• Daily

• Weekly

• Monthly

• Etc.

A timeframe can really be whatever you want. I personally trade systems on many different timeframes. You can also take a hybrid approach and have a system that calculates indicators on daily bars but only executes orders weekly or monthly; I do this as well.

Another thing to consider is which direction you plan to trade: long, short, or both? This is important because certain markets may not let you go short. For example, most stocks on the ASX exchange will not let you short the shares, but you could buy put options. On the TSX exchange, maybe 75% of stocks are shortable. Whereas most stocks in the USA stock market will let you go short.

With an equities portfolio, you may be able to get away with a long-only set of trading systems, considering equities tend to go up over time; though having some sort of short side hedge can be very helpful during bear markets. But with other market types, like forex or futures, the short side is more crucial because the fundamental drivers of price are completely different, and instruments can trend down just as easily as they can trend up.

Side:

• Long

• Short

• Both

Only once you have given serious thought to the items listed above should you move on to the next steps. This will give you direction in terms of what to focus on and prevent you from getting lost during the portfolio development process.

Some of these portfolio design options may be obvious; maybe you already know you want to trade equities long and short, but you have no idea what timeframe to focus on or what stock market(s). Maybe you thought trading intraday on the USA stock market was a good idea, but after reading this, you’re starting to rethink that decision.

Really do some research and think hard about the direction you want your portfolio to go, and remember you can always add a different set of market types, market regions, timeframes, and sides later on down the road.

---

The Portfolio Development Series Design Space:

For the portfolio that will be built in this portfolio development article series, we will focus on the following design space:

Market Type:

• Equities

Market Regions:

• USA Stock Market

• Canadian Stock Market (TSX)

• Australian Stock Market (ASX)

Timeframes:

• Daily

• Hybrid Weekly and/or Monthly

  • Look at daily data but make decisions weekly and monthly

Side:

• Long

  • USA, Australian, and Canada

• Short

  • USA stock market only as the Australian and Canadian are more difficult to short

Notice with this list, I am not completely boxed in. I have wiggle room to develop whatever systems I want within the bounds outlined above. I’m not boxing myself into USA large-caps or small-caps only, but rather just the USA stock market as a whole. I do not have any systems made yet, and I do not know where I will be able to develop the best system, so I am giving myself some creative freedom to look at a specific market region from many different angles.

Also, with the timeframes I have defined, I want daily, weekly, and/or monthly timeframes. But again, I do not know what timeframe will work best for the particular market, so I need some creative freedom. What is important is that I know not to look at intraday at all; those systems just won’t work for me. This helps me focus on one set of timeframes without getting distracted by others once I start the system development process.

This first step of defining the direction of portfolio design does not have to be overly specific; in fact, it is better if there is some wiggle room. What matters is that we took a massive design space and reduced it to a manageable region to focus on and gain expertise in.

---

Portfolio Performance Goals:

Now that the high-level direction of our portfolio has been determined, we know exactly what we need to focus on when it comes time to start building trading systems. But before we start brainstorming trading systems to develop, there is one more thing to consider, and it’s an important one: we need to define what our portfolio performance goals are.

Knowing these performance goals will help us figure out what we need to develop to get there. If the goal was to design a portfolio that had S&P 500-like returns with less volatility, that would be a significantly different portfolio of trading systems than a portfolio with the goal of the highest possible outright returns.

So first, here are some example questions to ask yourself to help you define the purpose of your systematic trading portfolio and what you want the outcome to look like. Thinking through questions like this will help you define your portfolio performance goals:

Why are you creating a systematic trading portfolio?

• Is it for income?

• Is it to support retirement later in life?

• Is it so you can quit your job?

What size drawdown makes you uncomfortable?

• If you woke up tomorrow and your portfolio was down 10%, how would you feel?

• What about 20%?

• What about 30%?

What amount of returns feel boring to you?

• If your portfolio only made 5% a year, would all this work be worth it to you?

• Where is the threshold in terms of returns that makes all this portfolio development work worth it?

• Remember, a portfolio is not set-and-forget once you develop all the systems. You’ll have to place the trades every day, and there will be ongoing maintenance and upkeep.

What length of time of being in drawdown makes you uncomfortable?

• If you were in a drawdown for 6 months, how would you feel?

• What about 1 year?

• What about 2 years?

What types of equity swings in your account value make you uncomfortable?

• Do you want slow and steady growth, or are you okay with fast but more erratic growth?

• If your portfolio was down 10% on Monday but up 15% on Friday, could you stomach this volatility?

• Or would you prefer being down 2% on Monday and up 3% on Friday?

These questions will help you get a general qualitative idea of what you wish to expect from your portfolio’s performance. Your portfolio equity curve is like a roller coaster with many ups and downs. Some roller coasters are small; they don’t go too high in the air, but they also don’t drop down as far. Others are huge; they quickly go up hundreds of feet and plummet down even faster.

So, the question is: how big of a roller coaster do you want to ride?

Once you understand why you want to develop a systematic trading portfolio and what its purpose is, we need to quantify some goals for the portfolio. We need some numerical metrics to compare our portfolio to so that we can ensure it falls within the bounds of our performance goals. Some great quantitative metrics to use are:

• Compound Annual Growth Rate (CAGR): The average growth of the account per year.

• Max Drawdown: The most the portfolio will be down at any given time.

• Average or Median Drawdown: The average/median amount the portfolio will be down at any given time.

• Max Drawdown Length: The maximum amount of time the portfolio will be in a drawdown.

• Ulcer Index: A measure of both the depth and length of the portfolio drawdown.

• Volatility: How much the portfolio equity wiggles up or down from day to day.

• Sharpe Ratio: A measure of portfolio returns in relation to the variability of those returns.

There are many other metrics, but these are some of the more common quantitative metrics used to evaluate portfolio performance.

An Example Of Converting Qualitative To Quantitative:

As an example, I’ll discuss my own qualitative situation and how I want my portfolio to behave and then convert those to quantitative metrics for the portfolio for measuring purposes. It will also show how I have chosen to structure my portfolio to suite my own goals. Use this example to help you plan your own portfolio metrics and design structure.

I trade systematically for the end goal of financial freedom one day. I do not want to work in the corporate world my entire life. Therefore, I need a portfolio that makes high returns so I can accumulate enough capital to quit my job or go to reduced hours sooner rather than later.

I want high returns but I don’t want to be pulling my hair out and constantly stressed to get there. I don’t want to feel the pit in my stomach of massive drawdowns or volatility. High and stable growth with reasonable volatility is the goal for me.

So, a CAGR of 20% or more is desired. This CAGR is reasonable for a systematic trading portfolio but also will help me compound my capital very fast. I am okay enduring some drawdowns along the way to hit that CAGR goal. I know from experience I can endure 10%–15% drawdowns with no problem, so if that was the average, that would be perfect. The drawdowns would start to get uncomfortable if they exceeded 25%, though.

So keeping worst case drawdown at 25% or lower, but closer to 10-15% on average would be ideal. As shown from part 1 in this portfolio development series, stacking a few momentum strategies on different markets can achieve this CAGR and drawdown goal. If you missed part 1, check it out here:

---

---

I’m also fine sitting in a drawdown for a decent period of time. A year-long drawdown would be manageable for me, but any longer would make me worried my portfolio is broken.

I know from experience momentum and trend following systems can have drawdowns longer than 1 year. So, to meet this goal, I’d need some mean reversion strategies in the portfolio as they tend to have shorter drawdown lengths. I could also implement some hedging strategies, which would reduce the drawdown depth, making them quicker to climb out of when the bull market comes back.

Finally, as far as equity curve volatility goes, 10% seems like a good goal to aim for. Statistically, a 10% portfolio volatility with a CAGR of 20% would result in a max drawdown of 20%–30% and an average drawdown of 10%–15%; which is right in line with my goals.

Finally, a Sharpe Ratio of 2 or greater I think would be pretty impressive, so let’s shoot for that. Sharpe is a good metric for risk adjusted returns. A lot of people like to use Calmar ratio (CAGR divided by max drawdown), and while it can be helpful, the denominator is only 1 data point. Sharpe is a better metric in my mind for risk adjusted returns as both the numerator and denominator take into account all equity curve data points.

I think these goals will help me accelerate portfolio growth while keeping drawdowns and volatility within a reasonable range. If you want more returns, you have to take more risk, so if I desired a higher CAGR, that would generally come with a larger drawdown and more volatility.

While a CAGR higher than 20% sounds nice, I don’t want to take on the extra risk required to get there. I don’t want to be drowning in stress because my portfolio is in a 35% drawdown. If you can handle the extra risk then no problem, take a little more risk and boost your returns. It’s all a give and take. So, to sum that up:

Portfolio Performance Goals:

• Portfolio CAGR Goal: >20%

• Portfolio Max Drawdown Goal: <25%

• Portfolio Max Drawdown Length Goal: <1 year

• Portfolio Average Drawdown Goal: <10%

• Portfolio Volatility Goal: <10%

• Portfolio Sharpe Goal: >2.0

After quitting my job in the future, I am okay living a modest life and living on the returns from my trading portfolio. My portfolio goals may shift when this happens. I will likely be less aggressive and want lower volatility and drawdown. I’ll adjust my portfolio when that time comes, but for now, I need the higher returns and will accept a little more volatility and drawdown as a side effect.

For this portfolio development series, the goals listed above will be used for the portfolio I am going to attempt to develop. We shall see if I can achieve them.

This portfolio development series is going to be a “develop in public” set of articles. I’m writing all these articles as I go. This article will be released before all the systems are developed and much before the final portfolio is assembled. This will give me opportunity to discuss any failures and talk about any required pivots as I go.

There is no guarantee the final portfolio will hit all these performance goals going forward. The best we can do is implement robust system development practices and do our best to be as diversified with non-correlated systems as possible. Outside of that, we can only extract what the markets give us.

The assumption here is that our backtested performance will be somewhat representative of the future, with some slight variation. Whether or not that is a good assumption, I don’t know—I cannot see the future. But we will position ourselves to achieve our goals in the best way we possibly can by developing a robust portfolio with multiple diversified trading systems on multiple markets and timeframes. That’s about the best we can do.

---

Benchmark Considerations:

While this isn’t the most important item to consider for portfolio development, it can help you determine if your final portfolio is worth trading. That item to consider is: