The answer is, pretty good! While past performance can’t be used to determine future performance, there has been an interesting shift from generally bullish forecasts, to the bearish outlook which was hinted at in June, and became fully flourished by August through to September.
The Table below tracks all 5-day forecasts, determining them as bullish or bearish even if the margin was just a 51% to 49% call. Only 50% forecasts (where no advantage can be determined), marked in blue, could no direction be determined and were excluded from the returns.
As the table shows, the 5-day forecast success rate was 66%. So even when the forecast margins were small, they were often enough to determine the outlook 5 days out.
The Average Return per trade was +0.54% for all trades (winners and losers combined). If bearish forecasts were ignored, then this return rose to +0.75%.
Splitting out the bullish and bearish forecasts; bullish forecasts were correct in 70.5% of cases, bearish forecasts in 42.8%.
With what looks to be a bear market in its early stages, it will be good to see how ChartDNA reacts to this change in market conditions.