MT4 971 compatible Tickstory Lite (free) vs Tickstory 1.8.1?

[Tickmate]

Hello o Tick Gurus

A simple question if I may please.

I am using MT4 build 971, and I have recently downloaded Tickstory Lite (the free version) and have been using this to put data into MT4 for backtesting. I have been doing some backtests – all seemed fine at first, but then I noticed that if I re-downloaded data for the same pair and ran a backtest with the same settings as previously, I got different results – very different. Like for example, twice the number of trades, etc.

But my main question is: Seeing that I am using MT4 build 971, is the data that I download using Tickstory Lite (free version) going to work at all? Or will it be corrupted and invalid since the MT4 971 build and Tickstory Lite free version are "not compatible" (or so I've read)?

Will I get markedly different backtest results if I pay for and download Tickstory version 1.8.1 and will all my backtesting problems thus be solved?

I have nothing against paying for the full version, but I would just like to get the official low-down on whether this will make my backtest results heaps more reliable, solve (invisible) compatibility issues between MT4 and Tickstory, etc?

Thank you very much for setting me straight on this.

[tickstory]

Hi Tick mate ,

According to our records you recently purchased so we assume that this message was created just prior. To answer your question, it is highly recommended that you regenerate your data with v1.8.1. This is because there may have been some useful changes along the way that are not currently available in the free version. With regard to differences in results, just make sure you are not coming across the issues mentioned in this thread:

viewtopic.php?f=4&t=1120

Apparently some users are having discrepancies after upgrading from Build 950 and you may need to follow up with Metaquotes if the listed solutions don't work for you.

Hope this helps.

[Tickmate]

Hi and thanks for your reply,

I have since realised the error of my ways. Using Tickstory 1.8.1, I have now seen 99.9% modelling quality backtests happening.

Before... I just **didn't know** you had to launch MT4 from with Tickstory! Et caetera... so I was probably just using history centre data, 5-year backtests went through in about 1.2 seconds, etc, and everything was *not working* even though I was totally unaware of this!!

Could you please let me know something, however? The makers of an EA I'm wanting to test have sent me the backtest results of a .set file I sent them, and their backtest says 90% tick modelling quality. Their results are substantially different from mine, even though some general features of the respective graphs can be seen to be related. I'm talking about... differences in profit factor from 1.18 (on their test) to 1.45 (on mine), vastly different profit level, etc......

So should these differences be seen as due to the difference between 90% and 99.9% modelling quality backtests?
Should I trust my 99.9% Tickstory backtests a lot more than their 90% backtests? (I don't yet know what their using for data).

Just trying to get to the bottom of this data and testing business, so that I can know that what I'm using is reliable.

Many thanks

[Tickmate]

Okay so I just found that my EA maker is using Alpari broker data for their backtests at 90%. Once again... should I be trusting Tickstory more than Alpari? Just an open question because I really don't know which way is up in such questions. Many thanks.

[tickstory]

Hi Tickmate,

The tick data that is exported by Tickstory should represent a closer simulation of what happens in a real market situation. In the case of Alpari/standard broker data, you are only getting a one-minute bars which will means that it is missing all the actual price movements that occur in a live environment. This will affect your ability to accurately assess your EA's performance - especially if it is a scalper. Of course, there are many caveats to back-testing in general and here is a thread that goes into more depth about this:

viewtopic.php?f=2&t=946

Regards.