To have the ability to import your own data in MT5 is it important for you ? - page 6

 

To all interested in this matter.


I have read a lot on Russian Forum, notably about this problem of tick data and possibility to import own data. There can be, and there are, endless debates between supporters of real ticks real and those who think that ticks calculated by the Strategy Tester are good enough. Between scalpers and those who think it is vain to hope to achieve a winning strategy on analysis finer than a minute, etc...

Personally I think this is a false debate, as a programmer and trader, I would have the opportunity to test and explore all possibilities imaginable for trading and developping programs. It seems quite natural that my expectations do not match that of Metaquotes which has its own goals and its own constraints. I am now convinced that it is not in the plans of Metaquotes, at least in the short and medium term, to give the opportunity to access actual ticks on MT5 plateform. Let alone be able to import data and use in the Strategy Tester. And it is a decision that can be understood if we start in place of Metaquotes, regardless of whether we agree or not.

I also think it is useless to fight against windmills and need to go ahead and try to find solutions. I want to support and implement, to the extent possible, the idea of ​​Ubzen to study the feasibility of a Virtual Strategy Tester. Rather than tinkering with his little gadget each in his corner, it seems to me that by making a joint project, there might be a way to do something very interesting.

EDIT : I point out that what I wrote here is only my opinion, I do not know the actual plans of Metaquotes.

 
angevoyageur:

To all interested in this matter.


I have read a lot on Russian Forum, notably about this problem of tick data and possibility to import own data. There can be, and there are, endless debates between supporters of real ticks real and those who think that ticks calculated by the Strategy Tester are good enough. Between scalpers and those who think it is vain to hope to achieve a winning strategy on analysis finer than a minute, etc...

Personally I think this is a false debate, as a programmer and trader, I would have the opportunity to test and explore all possibilities imaginable for trading and developping programs. It seems quite natural that my expectations do not match that of Metaquotes which has its own goals and its own constraints. I am now convinced that it is not in the plans of Metaquotes, at least in the short and medium term, to give the opportunity to access actual ticks on MT5 plateform. Let alone be able to import data and use in the Strategy Tester. And it is a decision that can be understood if we start in place of Metaquotes, regardless of whether we agree or not.

I also think it is useless to fight against windmills and need to go ahead and try to find solutions. I want to support and implement, to the extent possible, the idea of ​​Ubzen to study the feasibility of a Virtual Strategy Tester. Rather than tinkering with his little gadget each in his corner, it seems to me that by making a joint project, there might be a way to do something very interesting.

Those who don't understand the implications of synthetic ticks, or even understand or care what a tick is, are not likely to think that real tick data or even repeatable test data, in the form of history data, are important.  I wonder how many of the votes in this poll placed for "Would be useful" or "Irrelevant"  should really be placed for the missing choice of "What is a tick ?"

I am a Test Engineer,  I can't take seriously an environment where the test criteria and test data are outside of my control.  The History data are part of my test data.  To not allow the user the option of fixing the test data is, frankly, ridiculous and shows a lack of understanding on the part of Metaquotes.

 

Testing has to be Repeatable and Reproducible, without the facility to control our own History Data testing cannot be Repeatable or Reproducible. 

 
RaptorUK:

Those who don't understand the implications of synthetic ticks, or even understand or care what a tick is, are not likely to think that real tick data or even repeatable test data, in the form of history data, are important.  I wonder how many of the votes in this poll placed for "Would be useful" or "Irrelevant"  should really be placed for the missing choice of "What is a tick ?"

I am a Test Engineer,  I can't take seriously an environment where the test criteria and test data are outside of my control.  The History data are part of my test data.  To not allow the user the option of fixing the test data is, frankly, ridiculous and shows a lack of understanding on the part of Metaquotes.

 

Testing has to be Repeatable and Reproducible, without the facility to control our own History Data testing cannot be Repeatable or Reproducible. 

MetaQuotes concerns are not the same as yours.

You have several choices (non-exclusive) :

  • Use of the platform as it is and try to find solution for your concerns.
  • Use of the platform as it is and complaining about MQ for the changes you want.
  • Use another platform.

I'm sure the MT5 platform is the future, it has enormous qualities. So I chose option 1. What is certain is that MQ will not invest in changes that do not interest him, there is not ridiculous in that. I ask nothing better to deceive me about MQ plans.

 
Ubzen:


4>Q. How can I test an EA that makes Entry and Exit decisions based on tick information?

Unless mt5 guarantees it'll receive all Tick-Broadcast from the broker through CyberSpace, then I don't see how this could be called Accurate. I mean what would a tick-data trade logic look like? If (bid drops by 2_points) && (rise by 3_points) && {drops again by 2_points){ Buy!! } ?. If 2 people are trading this system and 1_person misses a tick, they're screwed.

I wanted to respond to this point as It's something I'm currently looking at . . .  if I miss a tick I'm not screwed,  I can know that I have missed a tick, or indeed more than one tick, and knowing I have missed a tick helps me not make an incorrect decision.  I'm currently missing an average of approx. 1 tick per M1 bar on GBPUSD,  this may well be Broker dependant.
 
RaptorUK:
I wanted to respond to this point as It's something I'm currently looking at . . .  if I miss a tick I'm not screwed,  I can know that I have missed a tick, or indeed more than one tick, and knowing I have missed a tick helps me not make an incorrect decision.  I'm currently missing an average of approx. 1 tick per M1 bar on GBPUSD,  this may well be Broker dependant.

what i am interested in, is how do you plan on profiting on tick stream, compared to time based stream, let's say S1 (one second)?

or did i misunderstand you? 

 
RaptorUK: I wanted to respond to this point as It's something I'm currently looking at . . .  if I miss a tick I'm not screwed,  I can know that I have missed a tick, or indeed more than one tick, and knowing I have missed a tick helps me not make an incorrect decision.  I'm currently missing an average of approx. 1 tick per M1 bar on GBPUSD,  this may well be Broker dependant.

You mentioned the buzz word "What is a tick?", so I decided to re-read the work of the contributers who came before us. Everyone had their definition of tick. Perhaps you'd like to tell the 66% (@current_votes) who voted non-critical what a tick is. If you miss a tick, it does-not necessarily mean that you missed a change in price. A tick can be as-large-as 100_pips during news time if the broker wanted.

Building a strategy on tick-price-action is very tricky business. I/MySelf/Personally decided a long-time ago to avoid building strategies around tick-data. But if you can make it work for you, then good for you. I just fail to see how someone missing 100_pips or just 2_pips could guarantee entering at the same price as another who didn't miss any ticks. One thing is for certain, the 2_people are not entering at the same time. I know for a fact, that you're not going to enter @ the same price unless it comes back down. So are you going to enter anyway because you missed a tick and forfeit the 2 missed pips or are you going to sit-out as the example intended? Either way, everything is not the same.

Testing a system which can overcome misses in price and/or direction is what I decided to focus on.

What is a TICK? - MQL4 forum
  • www.mql5.com
What is a TICK? - MQL4 forum
 
angevoyageur:

MetaQuotes concerns are not the same as yours.

You have several choices (non-exclusive) :

  • Use of the platform as it is and try to find solution for your concerns.
  • Use of the platform as it is and complaining about MQ for the changes you want.
  • Use another platform.

I'm sure the MT5 platform is the future, it has enormous qualities. So I chose option 1. What is certain is that MQ will not invest in changes that do not interest him, there is not ridiculous in that. I ask nothing better to deceive me about MQ plans.

Agree totally with this statement. Three days-ago I read the link you posted for the Russian forum. I'm embarrassed to say I actually read the entire thread on the spot using site_translate. Somethings get lost in translation however I think I comprehended most of it. Things I liked from the CEO was his frankness. He showed their experience, highlighted what they've done, why some things work and wouldn't work...etc. Things I didn't like was the "we know what's best for you attitude".

Like I've always said, meta-quotes gives me free software [actually paid for by brokers]. Until another platform decides doing that without trapping me unto a specific broker, I'll continue to use meta-trader. As to mt5 success, good-luck to them is all I can say.

 
Ubzen:

You mentioned the buzz word "What is a tick?", so I decided to re-read the work of the contributers who came before us. Everyone had their definition of tick. Perhaps you'd like to tell the 66% (@current_votes) who voted non-critical what a tick is. If you miss a tick, it does-not necessarily mean that you missed a change in price. A tick can be as-large-as 100_pips during news time if the broker wanted.

...
We have to verify that the same is true for MT5. I don't think that it was done.
 
Ubzen:

Agree totally with this statement. Three days-ago I read the link you posted for the Russian forum. I'm embarrassed to say I actually read the entire thread on the spot using site_translate. Somethings get lost in translation however I think I comprehended most of it. Things I liked from the CEO was his frankness. He showed their experience, highlighted what they've done, why some things work and wouldn't work...etc. Things I didn't like was the "we know what's best for you attitude".

Like I've always said, meta-quotes gives me free software [actually paid for by brokers]. Until another platform decides doing that without trapping me unto a specific broker, I'll continue to use meta-trader. As to mt5 success, good-luck to them is all I can say.

Yes, what I wrote here if mainly based on what I read on Russian forum where Renat (CEO of MQ) occurs very often. We must see the reality and move forward.
 
graziani:

what i am interested in, is how do you plan on profiting on tick stream, compared to time based stream, let's say S1 (one second)?

or did i misunderstand you? 

No,  you didn't misunderstand.  How I might profit is still a work in progress . . . if I come up with a workable theory,  and that results in a codable EA which I then code how will I back test it ?  well it won't be in the MT5 Strategy Tester.