Thread: Go/reverse

Hi:
I'm new to FX...
This attachment is a prototype...
Anyhow: the idea is: Put in any order into the market w/ TP & SL.
One of them will hit.
So, the momentary market direction is known.
So, enter an other order in the same direction.
If it exits w/ a TP: all is well.
If it exits w/ a SL, put in an order in the opposite direction.
& keep on...
(& hope that overall, there is more unidirectional movement, than "turning points".)
["turning points" just erase one gain]

The TP/SL size ratio could be varied....

The size of the TP & SL determine the TF.

On the attachment, I was on an H4 GBPUSD, but it actually trades (as an estimate) M5, but shows it on an H4 bar.
Well, that's it.

I have probs w/ coding, so the orderselect (on the prototype) might have to be made separately for each new ordersend [due to a reference to History].

(Still, the basic EA idea should be clear, from the post).
Thnx.

Originally Posted by bellco

Hi:
I'm new to FX...
This attachment is a prototype...
Anyhow: the idea is: Put in any order into the market w/ TP & SL.
One of them will hit.
So, the momentary market direction is known.
So, enter an other order in the same direction.
If it exits w/ a TP: all is well.
If it exits w/ a SL, put in an order in the opposite direction.
& keep on...
(& hope that overall, there is more unidirectional movement, than "turning points".)
["turning points" just erase one gain]

The TP/SL size ratio could be varied....

The size of the TP & SL determine the TF.

On the attachment, I was on an H4 GBPUSD, but it actually trades (as an estimate) M5, but shows it on an H4 bar.
Well, that's it.

I have probs w/ coding, so the orderselect (on the prototype) might have to be made separately for each new ordersend [due to a reference to History].

(Still, the basic EA idea should be clear, from the post).
Thnx.

EA doesn't work... order send error 130

Thanks

This is wierd...

Originally Posted by bellco

Hi:
I'm new to FX...
This attachment is a prototype...
Anyhow: the idea is: Put in any order into the market w/ TP & SL.
One of them will hit.
So, the momentary market direction is known.
So, enter an other order in the same direction.
If it exits w/ a TP: all is well.
If it exits w/ a SL, put in an order in the opposite direction.
& keep on...
(& hope that overall, there is more unidirectional movement, than "turning points".)
["turning points" just erase one gain]

The TP/SL size ratio could be varied....

The size of the TP & SL determine the TF.

On the attachment, I was on an H4 GBPUSD, but it actually trades (as an estimate) M5, but shows it on an H4 bar.
Well, that's it.

I have probs w/ coding, so the orderselect (on the prototype) might have to be made separately for each new ordersend [due to a reference to History].

(Still, the basic EA idea should be clear, from the post).
Thnx.

Such a system is called Digital system, which is not depending on any indicator, but only price action. So, why you are talking here about Time Frame?. Time Frame will change nothing.

I'm an expert in digital systems (or I - may be - I'm thinking so) and I already coded a lot of trading systems using such systems and also succesfully manage accounts using such techniques. My advice is to consider the lot size for the lost trades, but be careful, it is - sometimes - too danger.

Well, I'll look at Digital systems...
....I keep tring this & that.
Can't code too well.
Just found the "guppy" indicator... (A set of simultaneous moving averages, set at increments --- like 5, 7, 10, 15 bars; google can explain it better.)
But, it shows deviations from a "slow" baseline (like 45) as empty spaces..."bubble-s".
So, now trying to code trades that follow the trend by not having large bubbles. (Get ready to enter as prices cross the slow baseline, wait some # of bars to verify the "no large bubble". Exit as prices cross the slow base again.)
This should produce too few trades, but big.
But going multicurrency should keep the trading action fairly steady.
(& don't have too many orders out.)
Anyhow, I am a pennyless amateur, interested in trading (as it has such a large potential.)
My hoped for actual strategy is to save up my pennies [unlikely], & get a hold of HiRIDER from the "Forex robot world cup" site. To date: linear & no DD.

Hi X:
Yup, experts have been known to fade...
A possible solution is to move the EA to a new currency, that behaves the same way now, as when the EA was effective for, originally.
Alternately, the EA can adapt to its new environment (maybe).
Via hindsight: the GBPUSD between Dec 2000 and Dec 2009 had two regimes.
In one region, the bars on H4 alternated in size quite a bit, and the price stream oscillated about EMA 60.
Here the retracement strategy would have been overall profitable.
Ie: On any large change bar, bet against the direction of change, w/ a TP of about 3/4 the (H-L). (Pick a decent SL.)
This is my first picture.
In the second picture (chart scale is different, to show the whole of the trend), the bars are mostly the same size, and the price stream does not oscillate about the EMA 60. (Strategy: enter exit w/ the trend, wait some time for confirmation & condition tests.)
So, for the "adaptive" version: have the EA test to see which is its current enviornment, and use the corresponding strategy.
The test could be ongoing, & the EA could even have a library of environments/strategies.
But having the EA have an "auto shut off" (activated after losing x% from its highest equity) is ok, and it being manually transferred to a "well behaved" currency is simple. (Unless the EA itself monitors the various currencies.....)