Are "momentum" strategies possible? That is, catching a continuation of a strong move ... - page 10

 
Artyom Trishkin:

What was meant was this: if we have caught an impulse, and even opened successfully. Now we need to watch out that the flow of ticks has not slowed down, or "flipped". If there is a slowdown - stop at Breakeven, and watch further. Go further - wait for the slowdown. Caught the slowdown - we move the stop. If you see a stop or a reversal - it is better to close without making a stop and go into the mode of waiting for the impulse.

I did such a thing - it works fine, but you should fine-tune the trigger threshold both for the impulse itself and for a slowdown/stop/turn. And it's better not to catch more than two pulses in a row after a stop. Yes, the multicurrency unit is necessary here - because there are not enough impulses on one pair in a day. And before important news one should lower the thresholds.

I haven't finished this trading robot, in general - it was good on ticks history... But it's a story... I would like to finish it and see it in real life... Ehhh ... 24 hours a day is not enough.

You just read my mind - about impulses and catching them )) However I may add - slowing down does not mean stop. The pulse may continue - far away... It may bounce back just a bit - by a couple of points... For a reopening of an order you have a loss - in commission and on the spread. So it is too expensive to close orders without any need! That's why we need some more parameters!

Yes, 24 hours will not be enough! I have not even had time to formalize it all ))

If you don't mind, can you give me your algorithm for catching impulses?
 

Well not all is lost, looks like it's time to invent a time machine, then 24 hours is bullshit. (I've already given up on it myself, switched to opening prices and that's it)