I don't know actually what to make of the "predictions" of the backtesting engine. On both BTC-e and Gox I've managed to cook up the following (per exchange) settings:

	1wk	1m	3m	6m	1y
Gox	67%	347%	616%	435%	1739%
BTC-e	82%	352%	636%	424%	4063%

Am I just doing curve fitting? It would seem that the longer the time frame, the less effective curve fitting would be. I'm hesitant to believe the results could be even half this good, so what am I missing? Volatile times of course make for great opportunities...

I've traded alongside the bot now for a week, but I still don't trust it since this volatility seems to throw the EMA off, but maybe I'm looking at too small time frames. I guess I'm just looking at input from others so I could skip staring at the charts all day, trying to catch the waves

edit: While the bot turned a $1.1k investment in BTC-e to 1.4k in a week, in the same time on Gox manual trading yielded from 12k to 20k results (which is pretty bad considering how volatile the market's been), but I've not been sitting at the 'puter all day...