Hi,

I am wondering what the correct spread value to input into Zorro for backtesting the GER30 via FXCM?

Is it?

0.5*PIP; 1*PIP; 5*PIP;?

The reason being is that it can have a difference for overall profits particularly user smaller time frames when back-testing.

Thanks.

Bigger number is good for MT4, but I don't know for Zorro.

Anyway Did you backtest with tick mode?

Ok, thanks..

Yes, i have backtested using TICKS and my strategy is still profitable, but not as much as before, in particular it is not as robust over long back testing periods.

Do you always have run the backtest's in TICKS mode? Is this really necessary?

I don't mind working for another month over the evenings to overcome this problem to try and get the similar result as before..

What is wrong with the conditions being met and then waiting to enter or exit at the start or end of the bar period? If your running in live demo mode and TICKS are not set but the strategy seems profitable as it was in the backtest what is wrong with this?.

From the manual:

"If this flag is not set, an intra-bar price curve approximation is used for simulating entry and exit. In this approximation, a stop loss is always triggered before a profit or trail target, and trades closed by trade functions are sold at the open price of the next bar. This causes a less accurate, normally slightly pessimistic test"

Without TICKS my system performs better, so why change it?

I wonder if you could separate the entry and exits, and stop etc so one some can run on ticks mode and the other runs on bar period?

Cheers.

Rule of thumb: When your system performs different with and without TICKS, always test with TICKS. Z1 and Z2 use TICKS for the backtest.

TICKS can make a difference when your system often uses pending trades, or when it often exits a trade with a stop or takeprofit limit. TICKS won't make much difference with systems that mostly exit trades by script.

For the backtest, you need not set the Spread - the system does this automatically. You only need to set it when you want a particular spread for some reason.

Which is robust back testing with 99.9% tick data for MT4 and tick mode for Zorro?

The 99.9% testing in MT4 is mandatory because the last MT4 candle is tick based and thus normally incomplete. So without tick based testing, the MT4 backtest can be way off.

In Zorro the last candle is always complete, so tick based testing is not as necessary as in MT4. Still, it's better. You should use the TICKS flag under the conditions mentioned above.

Thanks for clearing up, now I understand.