Trading the DV Bounded indicator has been a very effective way to trade markets like the S&P 500 (or the SPY ETF) in the last few years. The basic method is to buy on a low reading (0.5).
As the summary table below shows, this simple method has worked fairly well across big cap, small cap and high tech markets.
Looking at the table, I wondered if there would be any advantage in trading the DVB, but switching the market that you apply it to based on a certain filter.
One useful trend filter is Frank Hassler’s Trend Strength Index which can be used in a number of ways to rank markets for their ‘trendiness’.
Trading DVB based on TSI Ranking
My first test was to see if you could improve returns by switching the market you trade the DVB indicator on based on the markets TSI ranking. The DVB tends to under perform in strongly trending markets, but more on this in a moment.
To keep it simple I looked at three equity markets. I’m sure others could be compared:
SPY: S&P 500 – Large Cap.
IWM: Russel 2000 – Small Cap.
QQQ: Nasdaq 100 – Tech.
The rotation is fairly simple, you trade the DVB indicator long and short only on the market with the highest TSI score.
I’m using the statistics from the new DV plugin. First lets look at the SPY trading DVB as a benchmark.
Now the rotation model:
We can see some small but significant returns, though some of the metrics show this comes with an increase in risk.
Is the TSI ranking telling us anything?
The first thing to note about the DVB indicator, like many indicators, is that it can’t work miracles with short trades in a bull market. The chart below is from trading the DVB indicator on the SPY, but only going short above, 0.5 and out of the market otherwise.
As you can see, this shorting method did very well around the credit crunch, but has done little more than tread water since then.
I wanted to see what a markets TSI rank might be telling us about the market mode in general.
The table below shows long trades only. The percentage is the average return per trade. The % are depressed because I was lazy and it is counting a no trade day as a zero return, so it’s best to look at comparative levels rather than absolute.
The top row shows all long only trades for that market. The rows below this what happens when a market has the top TSI with the individual columns showing how those individual markets behaving. E.g. when the IWM has the top TSI, the average trade score is 0.13% vs the benchmark of 0.06% of all IWM trades.
Interestingly, the best average returns for long side trades come when the IWM (Russell 2000 has the strongest TSI score). This may tap into the troops leading the generals phenomenon which runs that when small cap stocks are leading markets, it bodes well for all markets.
Some studies refute this, but I wonder if its because they measured small cap strength by simple out performance not by strength of the trend?
The short only trades table makes for the most interesting reading:
Short trades in general are below par when the IWM is showing the strongest trend. However, they are above par when the QQQ shows the strongest trend. Indeed, instead of trying to short the trend, you would be better off just buying and holding when the IWM is top instead of trying to trade the DVB.
Does this hint at a general switching of the gears?
E.g. if small caps show the best trend, it’s a broad based trend and you’ll struggle to make good money going short on any market. However, if those speculative, high tech stocks in the QQQ are showing the best trend, it might be a sign that markets are a bit frothy, with short trades about to become more appealing.
The average return is 0.13% (using my not accurate averages) buying and holding the IWM when it has the strongest TSI score. However, the average return is -0.01% if buying and holding the QQQ when it has the highest TSI score.
I’ll follow up soon applying these latter thoughts to the TSI rotation model to see if it improves things (It should).
Related posts:
