Well, they are going to put me on home schooling after this post. Here are two version of this I had running today. One with some rules trying to filter for over extended momentum and the other just the raw signals. I don't have it signalling weak buy or weak sells as M has mentioned are possible. The IWM is the raw crosses, you can see it doesn't back test well at the moment. The SPY is with a few filters and it avoided trading some good stuff today, and had losses the previous two days but it does back test well. I am going to use this coding and following here and watching live so I can make this my own. Just added that 3sma offset piece this afternoon, another good one!
here are spy trades with the complicated version of SPM <img src=http://elitetrader.com/vb/attachment.php?s=&postid=2147015 width=800>
Jahajee - I'm not exactly clear on your entry/exit rules. I think you may have explained this clearly in the other thread which was deleted. The entry is just the 2 min MACD/MACD avg cross? Or are you integrating something related to the oscillator you've got on the chart as well? I don't think you're using the 3sma +1 trigger. On th exit, are you using the KC or oscillator, or some combination thereof? If you're willing to post the EL code for the strategy, that'd be great, but if not, some more details to understand your entries and exits would be much appreciated. Sandy
See attached 2 minute ES Less spectacular results with Loose version of SPM - with crosses above or below zero - $340 profit per one-lot contract. Many trades are spurious - these can be filtered off with momentum. <img src=http://elitetrader.com/vb/attachment.php?s=&postid=2147027 width=800>
this doesn't back test as well, would be better to use this ninjascrpit as a big indicator and then trade them discretionary. For instance the ninjascript puts me in late at the end of a spike. I could usually fade in any moves after that for some better prices.
Yeah I use QT to trade with, and NT is like a tool I watch and see what is signalling. QT is rock solid and NT is fickle as a mofo and not truly reliable imho.
The main problem with automating a system is that your entries are based on mechanical signals, ie., most of your entries are late. We see the signals and take them in a span of approximately 10 seconds or less, 15 seconds is a little too long, when we're in the zone we can nail a trade in 5 seconds or less. You aren't going to get that type of performance with any retail level, off the shelf software. However advanced programmers may be able to achieve it if they know how to code and can use industry level software desinged for trading.
Quick question. Say you are looking for a long based on MACD. How do you use the SMA3+1 ? Are you looking for price to go below it, to get a good fill, then you go in ? Or is there something more to this. Thank you TV
Yeah that is a good point. I think it is good training for me, if I were to say this is a different style for me it would be an understatement. I do find that if I can code some things that are possible improvements, then I know that certain discretionary things are possible in trading this. In general I have on training wheels. I only got that last short in the true scalp style it seems you guys do and of course, it ripped down into the close lol! And afterhours!