What makes an ideal system? Like any investment, we can view systems in terms of the 3 key metrics of risk, reward and probability.  That is to say that risk (loss or drawdown), reward (gains or profit), and the probability of the risk versus the probability of the reward are assessed for each system.  

To evaluate a system simply based on its success is a very flawed approach.  One must look at both the odds of that success (how often do you achieve that success?) and the amount and odds of the failure (what is the worst case drawdown or how often does this system lose money and how much can it lose?).  

Let’s compare two systems’ results using risk-adjusted returns to better explain this critical point:

System A has won 8 of the last 10 years and compounded annual growth rate of 40% per year over the last 10 years.  Its worst drawdown was 5%.

System B has won 8 of the last 10 years and compounded annual growth rate of 50% per year over the last 10 years.  Its worst drawdown was 25%.  

In this example, system B would have made you 10% more profit per year.  However, with a drawdown of 25%, the risk taken for those additional returns is clearly far greater than the 10% extra profit achieved compared to System A’s 40% return.  I view this type of risk-reward analysis using a formula called the MAR Ratio. The formula is relatively simple to understand and it is used by professional investors all around the world:

System A’s MAR Ratio is 8 because we take the compound annual growth rate and divide it by the worst drawdown (.40/.05 = 8).

System B’s MAR Ratio is 2 (.50/.25=2).

Therefore, System A is clearly the preferred system.  However, the MAR Ratio is often used inappropriately because it doesn’t delve deep enough to evaluate the probability of that risk vs the probability of the reward.  If, for example, we viewed a system with a 1,000 year history and in that history the compound annual growth rate was 20% per year, and its worst drawdown was 20%, its MAR Ratio would be 1.  But in this example, over 1,000 years it had a positive return in 999 years without a drawdown, and a drawdown of 20% in just one year. This would clearly be a misleading MAR Ratio. 

So I like to look at the MAR Ratio but dive  deeper by analyzing the likelihood of those profits or losses, and that is what led me to a proprietary formula that helps to not only identify optimal systems by taking the MAR Ratio analysis several steps further, but also helps me calculate proper risk allocation and position sizing.  I call it The Score, and I will discuss that a bit later. For now, I want to stay on the point of what makes a great system.

So far I have talked about wanting to have an appropriate amount of ‘layers’ and a good Score.  However, there is more to discovering a great system. To be an ideal system, the frequency of the alerts must be within a reasonable range.  That is to say that I don’t want an excessive amount of trades in a season to outweigh the other systems being used in the overall portfolio or diminish the value of an individual trade within the system.  I also certainly don’t want it to generate so few trades that it becomes very exposed to the outcome of just a handful of trades to determine its success or failure. In general, the sweet spot quantity for trade alerts that a system will generate per season is typically 20-50.  There are some exceptions where a system can generate as little as 10 alerts or as many as 80 alerts in a season and the performance (risk-reward-probability) is so good that it justifies inclusion.

I also want the system to consistently win.  A baseline for this is targeting at least 80% of the years in the back-test group to be winning years.  This gives you a greater likelihood of finding a system that goes far enough outside of the standard deviation bell curve that it isn’t random success.  

The rest of the evaluation of a system comes down to some common sense assessment of the risk-reward-probability breakdown of the system.  I would be happy to take lower ROI if a system had minimal drawdowns and never had a losing back-tested year. This makes it a lower risk system so the returns can be lower than a more aggressive system and still fit nicely inside a portfolio of systems.

Using Systems to Create Non-Emotional Sports Trading

There are so many wonderful benefits to utilizing systems to trade sports.

1.   No Emotion – I have discussed this many times, but I can’t stress enough how this one element is a game-changer.  When you develop and implement a system you remove human input. The system dictates the trades to take, the amount to allocate, and when to stop using the system entirely.  The best traders in the world don’t fly by the seat of their pants. The money managers of the world’s biggest investment funds don’t trade on their gut. They systematically approach the market and the analytics, but most importantly, they do not let their emotion get involved. 

2.   Data-driven decision making – Sports have been on the forefront of data compilation and analysis for many years and that has culminated in what is perhaps the best data access of any tradable market in the world.  Data is analyzed over and over again using hundreds of different metrics and sources. They know everything from what a team does when the weather dips below 32 degrees to the win rate when the manager of a team wears a red shirt on a Tuesday.  The data is over-the-top and runs the gamut across the major sports, facilitating an unbiased review of each play, game, player, team, etc. This level of data-harnessing facilitates the creation of complex and historically back-tested systems that few other industries, if any, could rival.

3.   Set it and forget it – well this is a bit of a misnomer.  The reality is that the ScoreMetrics premium alert service is designed to make these systems a true set-it-and-forget-it system.  If you go the DIY route there is a lot of leg work in running the systems, but ultimately if you treat this like a business you can be relatively hands off with some minimal staff support.  Once the system is in place the rules are set, and the systems essentially run themselves. 

To accomplish this non-emotional systems-based approach to trading sports, the process is actually quite simple.  As you make your way through this book you will learn the secrets to developing proper systems. However, this is half the battle.  Luckily, the other half doesn’t actually require you to do anything, but rather to do nothing. You see, the point of the systems approach is to resist the urge to apply any human element.  The trick is to let the systems run autonomously, that is to say set it and forget it. The biggest mistake traders make is to enter a trade without a complete game plan, inevitably changing their plan as they go.  What our system allows you to do is manage a portfolio without having to micro manage individual trades. The rules are there for a reason, and the back-tested performance is only accurate if you follow those rules to the letter.