Real and historical graphs are different.. spread?

[fdanilo]

hello to all
Meanwhile, I also thank you for the very useful program.

I have a doubt.

There is a difference of a few pips between the chart in online mode and the chart downloaded from the historic.

I download the historian with the time on my broker and I start MT4 from tickstory in offline mode:

As you can see the day 2013/09/11
has
Open 1.3267
High 1.3324
Low 1.3243
Close 1.3313

Equal in backtest.

But if I put the MT4 online, this is what I see:

the same day 2013/09/11
has
Open 1.3267
High 1.3323
Low 1.3243
Close 1.3309

and even if I use the tester adding 2 pips of spread, it's the same
graphic used for the backtest is different from the real online

even if they are few pips, the results will be different, especially if use smaller timeframe

I something wrong?

if, once I have downloaded the historian, I login in MT4 with my real account and update the graph, all the data is correct, he updates the spread, but..... I'm updating all the historian, and this is not good, right?

[fdanilo]

I'm noticing that the problem may be given by gmt

I'm trying all the gmt for different brokers, but there are many differences from the 4H chart on
I tried it with Pepperstone, FXDD, Alpari
all GMT +2, +3 for each broker, I also asked the assistance to be precise
but do not coincide GMT in historians with the downloaded graphics in real time
if they coincide are only for the summer, with daylight saving time change yet

can someone help me?
which broker do you use and which gmt?

thanks

[fdanilo]

No one can help me??

I can not get graphics equally between historical offline and real online
even if I follow all the tickstory setup instructions

please....

[tickstory]

Hi fdanilo,

I'm not sure if I understood correctly - are you comparing the historical data (which is sourced from Dukascopy) to that of your broker? If so, there is always going to be a small difference given that each has their own liquidity.

Hope this helps.

[fdanilo]

hello, I try to explain better .. I'm using google translater.
I would like to do backtest on graphs as identical as possible to charts of my broker.
For example if I do backtest on historical metaquotes and then I used my EA on demo metaquotes, I know that the graphs correspond exactly to the historical (spreads, gmt, summertime, etc. ..).
So there is more chance that the optimization is valid.
Instead downloading the historian with Tickstory, also if I configure that correctly and I try more gmt, I can not find the historical identical to my broker.
Or the candles correspond only in summer and not in winter, or the 4 hours candles have different time..
There is always a difference of at least 4-5 pips for each spark plug.
With hundreds of different candles is logical that the indicators are very different, so the optimizations.
I guess I do not make mistakes, I am careful to keep the mt4 offline, I compare the historical graph offline with another mt4 online.
Another attempt that I will do tomorrow is to use the mt4 of "Fair Trade Technologies" that is connected to Dukascopy, I see so if historians downloaded from Tickstory (with Dukascopy database) are identical to charts FTT.

There's something I wrong? Can you help me figure out if they are in the right direction or something wrong?

Thank you very much.

[tickstory]

Hi Fdanilo,

Regarding the broker spreads, you will always get some sort of difference between different brokers. Remember also that Tickstory is exporting a fixed spread at the moment so just bear that in mind.

With regard to time-zones, do you know what time-zone your broker is working under? Do they shift time for daylight savings? You will need to pick a matching time-zone from the list - when doing so, it indicates whether the chosen time-zone will automatically adjust for daylight savings. Is it only during summer-time that the issue occurs?

If you are still having issues after trying this, please let us know a specific day where you are having issues so we can look at this further. In this case, you will need to give us the charts of what you see on your broker's chart to see why they don't match up.

Thanks.

[fdanilo]

hello
to the left you see the graphic downloaded from tickstory with mt4 offline and to the right the online graph

in this picture you see that the graphs of the two mt4 are identical, because we are in summertime

in the second image, we are on the same graph but in the month of March and how do you see gmt changes, and consequently can also change the candles and also the results of the backtest.

I've tried dozens of decisions and gmt different, but no one is always equal to the graph online

have I explained myself better?

thank you very much again

Danilo

[tickstory]

Hi Fdanilo,

I assume those online graphs were from Fair Trading Technologies? Based on their website, their timezone is set to GMT + 1, and it also seems like they do not take into account daylight savings. Please try to export using one of the UTC + 1 timezones that does not adjust for daylight savings and let us know if this fixes the issue.

Regards

[fdanilo]

Hi, it's Pepperstone
you can see in first image
I've tried many GMT.. wich do you recommend?
thanks

[tickstory]

Hi Fdanilo,

We have done some further testing including creating a "pure" time-zone shift of +1 hour and it still doesn't match with what your broker is showing.
Do you have the specific minute that the two graphs go out-of-sync? Are you able to post a screenshot of this?

Thanks.