Chinook's EUR/USD (E/$)Mumblings
- Thread starter chinook
- Start date
First of all, I tested it for US session hours. New entries would be made only between 8:40-14:30 EST . If a position is open beyond 14:30 EST , it would be trailed until CME close or getting stopped out.
In terms of money management, it will not take new trades after about 25 ticks total loss. There's no limit on # of trades but it usually trades max 3-4 times a day.
The problem is I only took long trades. I'm trying to figure out a way to include volume in the direction bias. Or I'll couple it with the daily chart. If the trend is up/down in the daily, it'll only take long/short trades. I think this'll be the smart way to go. Anyway, intraday volume is secondary to the daily trend.
How about "seeding" the system with directional bias ? As far as I can remember you have had a correct directional bias more often than not.
Directional bias is quite difficult to get right at any given point of the day. Also, the currencies have a lot of "events" that are even more difficult to model into the system.
I see that you are going to use the mechanical system to circumvent the "human factor" of your system trading, but the human "nose" or "bias" could still prove useful - only it should probably not be allowed to be changed or reset during the day ... which is the usual day-killer for you I think.
You will probably agree that if you do some really thorough preparations reading analysis, news and study a lot of overnight changes on a broad currency basis (i.e a lot of pairs and perhaps even fixed income, commodities - whatever is on people's minds - the popular talk) and then - only then - seed the mechanical system and let it start it's work, well, then you will accept whatever losses are incurred over that day as part of the system and showing that the money management works - and doesn't get hampered by "human intervention".
Your "human part" would mean preparing as best you can before starting to trade the system, that would be an important responsibility, and bring confidence in both your bias and seeing if the mechanical system is pulling it's worth or need adjustment. You would not be very responsible if you just wake up and then start the mechanical system immediately, how would you be able to judge it's performance if you did not thoroughly prepare ?
You would get a much better objective basis for judging the performance of the system. Some days the bias might be right - but some event or change might tilt the price levels - still that should not change how your system is biased since most times the general bias prevails on the day. Agree on any of this ?
The next post will probably set off the 500-page jubileum ...
Quote from Gringinho:
How about "seeding" the system with directional bias ? As far as I can remember you have had a correct directional bias more often than not.
Yes, this is exactly what I'm thinking about. Looking at the volume and other factors important in the day. I'll tell the system to go long or short only. I'll also be able to change the directional bias. The important part is that entries and stops will be 100% mechanical. This should cut the impulsive part.
However, I can't backtest this--I mean changing the bias for every day. But overall, the system will make money as long as we have smooth semi-directional days.
I agree. I'm usually good with the overall bias but the occasional impulsive entries take a toll.
This is also my global idea. I can turn the long/short switch of the system and accept the results.
Your "human part" would mean preparing as best you can before starting to trade the system, that would be an important responsibility, and bring confidence in both your bias and seeing if the mechanical system is pulling it's worth or need adjustment.
You would get a much better objective basis for judging the performance of the system. Some days the bias might be right - but some event or change might tilt the price levels - still that should not change how your system is biased since most times the general bias prevails on the day. Agree on any of this ?
The next post will probably set off the 500-page jubileum ...
Yes. I agree. However, as I mentioned earlier it's impossible to test this. But I can test the system as long or short only two seperate systems.
Actually, you know it's all in the money management. Of course, it helps to get the direction right but the real key is position sizing, stops, scaling-in and out profit targets etc.. This is why, I might just make things really simple for myself and get the directional bias from the daily chart trend. And seed that into the intraday system.
Yes, and that would of course help in judging the performance of the system - the seeding bit I think.
Just now I think there are a lot of nice long USD again - I just shorted on high EURUSD. It's a scalp for now - with profitable adaptive tight stops.
Contrarian scalper as always.
edit: stopped at +7 - first trade of the day.
Can I ask, why do you compare the December contract with the March contract, there are no January and February contracts? Or is there a special reason for this?
Maybe a test of 1.33 is possible, after which the eur will slide down again?
Maybe a test of 1.33 is possible, after which the eur will slide down again?
I watch for short-short term differences - it's like tape-reading with double input - double clues. The single contract is less noisy itself as frequent trades are spread out on both contracts - making direction perhaps be more intelligble. I do this for 6J, not the 6E right now.
With 6E, I studied the charts - then the book and saw high 1.3290s on the march contract as "very high" and ready to reverse. I only got the +7 instead of the roughly 20 it has fallen, but I'm playing careful today.