There were no position changes in the Country models this week. The Basic model had another selloff with TMF continuing to decline dropping below its 50 DMA on the Friday close. The Stops & Targets model actually put in a positive week but lagged behind its benchmark.
The broader market had another strong week with some good news coming out about Greece and Russia. The SPY put in new all-time high closes on Thursday and Friday.
This Week’s Strategy Lesson: ETF Model Drawdowns (Part 2)
In last week’s article we began discussing drawdowns. We looked at how to define them and how understanding their characteristics can help us determine how to trade a system. In this week’s article we will look more closely at some of the data we have and what it can tell us about the likelihood of encountering drawdowns of various severities.
Getting Started
One of the biggest decisions related to the ETF models is how you get started trading them. A lot of different factors can go into this decision, like how much capital to deploy or if you should “leg” in slowly or leap in head first.
One thing that can give traders pause is concern about initial drawdowns. We have all experienced the displeasure of entering a trade only to find out that we bought at the high.
Drawdowns are never fun, but they are often easier to stomach when we have already made money on a trade, essentially “playing with the house’s money.”
Last week we looked at the maximum historical drawdown of each of the models. But it is important to note that many of these drawdowns occurred after a period of strong gains and so far the models have always gone on to put in new equity highs afterwards.
As a trader of these systems, you really only encounter the maximum pain point for these drawdowns if you were “unlucky” enough to get started on the day or week the drawdown starts. Which brings us to another way to approach the question of drawdowns.
What if instead of asking what the worst drawdown was, you instead asked the question, “If I started trading this model now, what is the likelihood that I will immediately experience a drawdown in my starting capital of a certain size?” Once we know the probability and the size of potential drawdowns, we can then use a position sizing methodology so that we can comfortably handle a drawdown should it occur.
Drawdowns and Probabilities
The ETF Complete Portfolio is perhaps the easiest model for us to analyze in this method, helped a lot by having a very mild 13% maximum historical drawdown. It goes without saying that if you had pegged a 15% drawdown as your “pain point,” historically you could have started trading the Complete model at the worst possible time and still never encountered a 15% drawdown.
However, we can dig deeper and look at a couple of other scenarios. If we looked at a 10% drawdown level, we find that only 2.1% of all the possible weeks you could have historically started trading would have led to a 10% drawdown from your initial starting capital (drawdowns here are based on weekly closes).
Of course, the smaller the drawdown the more likely it will occur. A 5% initial capital drawdown occurs about 24%--but when you get into drawdowns that are this small, you start dealing with the normal “noise” or volatility of the model.
The Sector model tends to have the highest returns, volatility, and drawdowns, so we can expect a very different picture when running the same analysis on this model.
We already know that the maximum drawdown for the Basic version is right around 32%. But of the 425 weeks the model has been tracked, you would have had to start trading exactly on July 31st, 2009 to actually experience the full drawdown. Even a 25% drawdown would have been very rare at 2.6% probability.
It is only once you get into much smaller drawdowns that the probability starts to rise. A 15% drawdown from initial capital is about 14% likely and 10% drawdown from initial capital occurs about 33% of the time. The Stops & Targets version has still smaller probabilities of the same size drawdowns.
The Country model has a maximum week-ending drawdown of just over 23%, however, similar to the sector model, to experience the full drawdown from your starting capital, you would have had to start trading the model on exactly May 4th, 2007.
Looking at the same 15% drawdown the odds are around 6% that you will experience that initial drawdown and the 10% drawdown is right around 14% probability.
Risks and Position Sizing
We believe that ETF momentum-based models offer great opportunity for outsized long-term gains. But one of the surest ways to miss out on the return possibilities is to never get started trading or to be “spooked out” of your positions by a quick drawdown.
Now that we know a little bit more about the risks associated with trading the models, we can use this information to help us determine how much trading capital to put at risk.
For instance, if you were to start trading the Sector Stops & Targets model and decided that you didn’t want to risk more losing more than say $1,500. Looking at the data we have provided, the $1,500 translate to 15% of $10,000 and we would know historically, there is about an 8% chance that you would reach that limit.
Depending on which model you are trading, your risk preferences, and how much capital you could potentially invest, you can now make an informed decision about how much capital to deploy.
It should be remembered that the data here is based on the over eight year historical track record of each model. However, the future isn’t always like the past. The future could include more or less favorable trading conditions.
One way we can be better prepared for future uncertainties is through running Monte Carlo simulations on our models. Next week will continue this series starting by explaining how a Monte Carlo simulation works and seeing what there is to learn from this type of analysis.
The Current Condition of the Model
For the country model, we are in FXI, IFN, and TMF. FXI is currently in fourth place but it will take another decent move down in FXI or up in SSO before a position change will be triggered.
Stay tuned to daily updates for any position changes.
Here is a summary of the weekly performance of all the ETFs that the strategy monitors.
Best wishes for your trading,
James Kimball
Trader & Analyst
MarketGauge
