ETF Complete Strategy Insights: How to Follow the ETF Models (Part 5)

James Kimball | June 23, 2019

The ETF Complete model closed the week up +2.1% compared to the SPY which closed up +1.6%.

After the Federal Reserve reaffirmed its earlier guidance and accommodative willingness to cut rates if needed (at least one rate cuts seems likely this year), the market edged out a new all time high. Many issues remain with trade and tariffs, recent weakness in economic data, and rising tensions with Iran being chief among them.

Stay tuned for the daily updates and log into the website to see holdings and additional performance data.


This Week's Strategy Lesson: How to Follow the ETF Models (Part 5)


This week we are going to wrap up this series looking at how and when to start trading the ETF models. As we saw last week, the ETF models have high odds of positive six and 12 month returns, even when the starting periods are chosen at random. For six months out, we were seeing values in the mid seventy percent range and for 12 months out, the odds of positive returns were in the high eighty percent range.

However, we all know that not every market is the same and it would only seem obvious that markets and the models might behave differently when putting in new lows than when putting in new highs.

We have the dueling phenomena of “Trend Persistence” and “Mean Reversion” fighting it out on any market move. Trend persistence refers to the tendency for a move in the market to continue as the underlying factors that prompted the move continue to play out. Mean reversion is the idea that you are more likely to see a more normal outcome following an extreme outcome or in markets, that prices will tend to revert back to their average.

When you have a year like 2013, when the market was up over 30%, trend persistence was the more dominant effect. Mean reversion is more evident in years where the market can’t pick a direction or any sharp move is closely followed by a retracement.


The table above looks at each instrument or model when it is at or within 5% of its six-month high and looks at the odds of a positive six-month return going forward, the average of that return, and the “benchmark” six-month return (which includes all periods regardless of market conditions).

The odds of a positive return in this condition remain high, around the numbers we were seeing for all periods, while the six-month return percentages were also close to their benchmark. This doesn’t mean that there aren’t any “bad” times to enter the models as we know there are retracements and temporary highs. However, from a performance standpoint, there doesn’t seem to be any reason to adjust or avoid entering the models merely because they are at or near new highs.


When looking at the flipside of these statistics (markets trading near their short-term lows), we see some more dramatic changes in performance, particularly for the SPY. Over our sample period from 2007 to 2017, the average SPY six-month performance was 3.3% versus 0.2% when its trading near its lows. The steep and protracted decline in 2007 through March 2009 was responsible for pushing this statistic lower than it would have otherwise been in the last few years. If we excluded that period, we would see significantly more mean reversion in the data from the lows as most of the market corrections since then were shorter and, at least in hindsight, turned out to be good buying opportunities.

Overall, you certainly can make a stronger case for caution when markets and the models are near their lows, though we still see generally positive outcomes and averages don’t always tell the whole story. Also, not all models spent the same amount of time in these high or low states with significant differences in the number of observations for each.

These statistics back up the idea that you don’t have to rely on perfect market timing or luck to profitably trade the ETF models. The inclusion of short and alternative instruments help balance the portfolio when markets are correcting, while the strength and persistence of trends can help keep the engines running when at new highs.

Hopefully you found this series helpful in starting and managing any of these ETF portfolios. If you ever have any questions about any of these topics or others, feel free to contact us via email or messaging or you can ask questions in our live monthly coaching sessions.