Should You Include ETFs in Your Trading Strategy?
By Dave Mabe
Several traders have asked me this question, so I thought I would share my answer with the group.
The short answer: a reasonable, default approach is to exclude them from your strategy.
The longer answer:
You would expect individual stocks to have stronger signals than ETFs, which represent multiple stocks.
Counterpoint: You might expect moves in ETFs to be less prone to manipulation than with stocks.
We can all agree that ETFs are fundamentally different than stocks, so treating them the same is likely a mistake.
It always pays to look at the backtest, though.
You can always add a column for is_etf and then filter that column in the Cruncher to run reports for:
Including them in your backtest
Excluding them from your backtest
Excluding trades that aren't from ETFs
Whatever the data suggest, philosophically, it feels like looking for unusual moves that define your strategy HAVE to be diluted in ETFs.
Excluding them is a reasonable default since you avoid having to consider thorny questions like:
Does it make sense if you're long stock XYZ and also long (or even short) the sector ETF that XYZ belongs to?
What if you're long NVDA and also short NVDD? (This situation will be more common than you think.)
Now you've effectively doubled your exposure in one symbol (but your strategy and optimizations are completely naive to this).
For all these reasons, excluding them is simpler.
I can imagine strategies where including them would make sense, but they're relatively rare.
Other approaches to think about:
Exclude single-symbol ETFs
Exclude leveraged ETFs
Exclude sector ETFs
Tomorrow I'll share practical ways to identify and exclude ETFs from your backtests and real-time signals.
-Dave
P.S. Are you in a drawdown and ready to quit? Join the other traders who reversed their drawdowns and are printing money with MabeKit.