Help me write a inner bar EA | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

3 Bar Fractal with mid-bar and bar on each side 11 replies

The Inner Game of Trading 101 replies

New Renko Bar and Reversal Renko Bar Indicator Alert Help Please 5 replies

Pin bar, pin bar, pin bar 0 replies

Make Friends with Your Inner Interpreter, by Dr. Van K Tharp 0 replies

Help me write a inner bar EA

  • #1
  • First Post: Jan 16, 2009 11:51am Jan 16, 2009 11:51am
hi guys, i'm trying out for the 1st time writing an EA of my own. it's based on a system i found on another forum. i picked it because it was an extremely easy system (only 2 things to look for), but backtesting shows craptistic results.

here are the rules of the system:
1) find inner bar
2) analyse stochastic(8,3,3)

long: if stochastic < 20 and going up, enter at 2 pips above high of inner bar. stop loss at low of prior bar of inner bar.

short: if stochastic > 80 and going down, enter at 2 pips under low of inner bar. stop loss at high of prior bar of inner bar.

seems simple enough right?

here is my main() function:

Inserted Code
   // find inner bar
   if (High[2] > High[1] && Low[2] < Low[1]) {      
      // check stochastics
      double sto_main = iStochastic(Symbol(), NULL, 8, 3, 3, MODE_SMA, 0, MODE_MAIN, 0);
      double sto_sig = iStochastic(Symbol(), NULL, 8, 3, 3, MODE_SMA, 0, MODE_SIGNAL, 0);

      if (sto_main > sto_sig) { // up trend
            if (sto_main < 20) { // strong signal only
               double stoploss = Low[2] - (2 * Point);
               double takeprofit = Ask + (40 * Point); // arbitrary take profit
               
               if (!AlreadyBuyOrder())
                   OrderSend(Symbol(), OP_BUY, 0.01, Ask, 0, stoploss, takeprofit);
            }
         }

alright, i suppose forex wouldn't be hard if an EA as simple as that is a winner. i'm only going for long positions as you can see. but past year testing on EUR/USD, i only won 3.6% of trades.

shouldn't the oversold signal from the stochastic at least give me a higher win percentage?
  • #2
  • Jan 16, 2009 12:06pm Jan 16, 2009 12:06pm
You might need a clearer take-profit strategy. It's not going to help with the win percentage you posted, but it can at least start you in the right direction.

Also, with a system like this, the period you use will be CRITICAL. Make sure that you follow whatever the recommendation is from the site you got this from. A system like this will do much better in the longer time periods (at least D1)
  • #3
  • Jan 16, 2009 12:19pm Jan 16, 2009 12:19pm
thanks for the quick reply!
i've tried on higher timeframes and results don't appear to be better.

there's a key part of the system that i'm not following and that is trading against the main trend. you make trades on the 15MIN, but only if it is in the same direction as the trend on the 4HR.

i modified it to include cross-over of MAs from the 4HR.

here's the updated code:
Inserted Code
int start() {
   // find inner bar
   if (High[2] > High[1] && Low[2] < Low[1]) {      
      // check stochastics
      double sto_main = iStochastic(Symbol(), 15, 8, 3, 3, MODE_SMA, 0, MODE_MAIN, 0);
      double sto_sig = iStochastic(Symbol(), 15, 8, 3, 3, MODE_SMA, 0, MODE_SIGNAL, 0);

      if (sto_main > sto_sig) { // up trend
         if (sto_main < 20) { // strong signal only
            double stoploss = Low[2] - (2 * Point);
            double takeprofit = Ask + (40 * Point); // arbitrary take profit
            
            if (iMA(Symbol(), 240, 5, 0, MODE_SMA, PRICE_CLOSE, 0) > iMA(Symbol(), 240, 10, 0, MODE_SMA, PRICE_CLOSE, 0)) {
               if (!AlreadyBuyOrder()) {
                  OrderSend(Symbol(), OP_BUY, 0.01, Ask, 0, stoploss, takeprofit);
               }
            }
         }
      }
   }
   return(0);
}

unfortunately, it only improves it to 25% win.
  • #4
  • Jan 16, 2009 12:27pm Jan 16, 2009 12:27pm
Quoting hyperbling
unfortunately, it only improves it to 25% win.
Thats not a bad win as long as you're looking at 4:1 returns. Why dont you adjust the take profit? Also, you're stop seems to be awfully small for a D1 type chart. That would stop out quite quickly, quite often. Where did you pull that from?
  • #5
  • Jan 16, 2009 12:30pm Jan 16, 2009 12:30pm
http://forums.babypips.com/free-fore...s-per-day.html
  • #6
  • Last Post: Jan 16, 2009 12:39pm Jan 16, 2009 12:39pm
My guess is that you're just getting stopped out too often. 2 pips below the previous bar basically means that if you're pair doesn't go up immediately by 40 pips then your trade will lose. Try a "looser" stop.
0 traders viewing now