I came across this example code (Example 2: SPY Constituents in an Uptrend) on the site and I don't understand how this very basic, but logical - - price above SMA, beats the reference ETF hands down, over longer periods.  Only diverges during volatile periods, but quickly recovers after that, ultimately beating the EFT hands down.

The only changes on my part was to use 50d SMA in place of the 200d SMA and rebalance daily.  All fees are included so the high turnover should not be an issue.  Is there a bias leaking somewhere???  This seem too good to be true.