Alex is correct, it should be possible to backtest the XTL, as the indicator doesn't change historical results based on current action (as the wave count part of aget does).
I don't think the results of such a backtest will be impressive, as in my experience XTL works best as part of a system, and not as a system all in itself.
Go to the eSignal University section and look for the "Catch 22" paper in the Get section. It is a report on using XTL with some detail on the author's method. It has results rom his testing. There is also an older message thread about the paper.
Comment