the main problem following trend by EMA lines is that lot of times, you end up losing bitcoins due the fact EMA lines
have a "lag" showing the trend because they need strong change values to cross themself
so the single way you make monye here is when trend are long and with high value discrepancy,
you can't take advantage of sudden drop values, thats why i think the bot needs a better way to manage buy/sells when EMA indicator give an alert.
for example, at 22 Trades, in 2 months, the bot managed to "win" a 593%,
this "win" is just thanks to the fact bitcoin price rised a lot,
but at 22 negative trends there are 14 of them that return negative balance, so those negative trends were not profitable due the way
EMA lines work themself,
for example, day
19-11-2013 at 12:24- there was a sell at 600$ following a drop trend
the price hit around 450 or maybe less, thats a 25% price drop,
and what did bot do? well... same that it did in those 13 oportunities, just wait until follow the trend, and from 414% profit, turned to 413%
i know its short lose, and the bot is not trying to squeeze all single drop oporunities due the fact there is no trading engine, just indicator follower
if the non greddy system explained in my last post was triggered, just at 10% drop, would had been enough to get an nice extra profit there,
if from 22 trades, 14 are losed trades, means 63% of the time it catches a dropping trend, was 0% squezzed, it just cut off loses buying back when the trend rises again,
this scenario happened 63% of the trades in 2 months, and this means 63% of the "buy cheap" oportunities was lost and the sentence theory "sell high, buy low" was not profitable at those point.
With simple personalized rules for trigger buy events on a % market price drop, the profit would had been much higher and the drop trend oportunities much better squeezed,
Done arguing, the math to me is pretty easy. And since I'm up a few hundred bucks over the last few days, I'm good.
.