Short-Sell Backtesting with ETFs

 If you would like to test Short-Selling on ETFreplay, use the versatile app called Rel Str - Combine Portfolios.

You can test going short the top or bottom ranking securities in a list.

A Backtest Example For Inspiration: EFA vs QQQ regime

This backtest defines a Regime by comparing the performance of EFA and QQQ, two standard ETFs with plenty of market cap. The backtest then decides to allocate to EFA or QQQ depending on which Regime is in place. SPY is held in either case as a core position:


Low Drawdowns and High Returns. 2017 thus far has been investment nirvana

Over the past ~15 years, the bond market has generally had positive single-digit returns and also single-digit calendar year drawdowns.  As the gentlemens asset class, bond ETFs generally don't have the anguish associated with the big drawdowns of many equity ETFs.

For a reference point, below is a snapshot of Calendar Year returns and drawdowns for LQD, an investment grade bond ETF:



What is remarkable about this year is the combination of high returns with extremely low drawdown in some traditionally high-vol, high-drawdown segments - such as emerging markets.   2017's max drawdown for emerging markets has actually been less than most BOND market years.



+27% YTD total return near the end of the 3rd quarter of 2017 vs just -3.5% drawdown.   Obviously, strong return and Low volatility leads to high rankings in our Relative Strength models.  Uptrends can have some violent short-lived corrections but investors can manage such volatility by tilting their portfolios away from the weakest segments.

Link to the tool in this blog (subscribers): ETF Max Drawdown

Does an ETF track its underlying index by its price?

Does the ETF market price track the ETF's underlying index?

No.    You must calculate the Total Return and use that resulting data series for accurate backtesting signals.

ETFs as you may be aware are designed to track an index.  In order to have the ETF track the index in terms of a backtest, you need to re-vinvest the dividends and distributions paid.   An 'index' of course doesn't make distributions since it is not an actual investment product.  So the 'index price' actively builds the dividends back into the calculation of the index value (price).   But ETFs don't do this, they must pay out distributions by SEC law.   

Thus, only the CALCULATED total return data series can represent the INDEX PRICE (the ETF PRICE does not).   If you just used the ETF price, you will get inaccurate signals.  You can observe the difference between the price return and total return with our Total Return vs Price Return tool.