I know i read a post whish says "data is same in live than in tester mode ",
however i see difference of 0.5 pips lot of time :

i did a strategy whish record close bars value each 15s in a file in live account,
after this i i run this one in tester mode using history data,
and i clearly often see a max difference of 0.5 spread.

The problem is that i use a RSI indicateur, as it builds with multiplication of close bars value,
so the result is lot of time significativly different from live to tester mode,
By this way, my strategy results is very difference from live to tester.

What can i do ?

Thank you.

Dear sir,

First please make sure that you are comparing the same data in the tester than you do in live. As per your wording,
it looks like you are recording the close value every fifteen seconds. So, please be so kind to check the following :

1. What is the method you employ to record and store the close value every fifteen seconds ? According to the answer, are you sure that the strategy is doing the same in the tester mode ?
2. On which reference value is the close value recorded, as well as the indicator ? Is this bid, ask or some other ? And, does the strategy use the same reference value in the tester mode ?
3. Finally, asking the time of the tick is the best way to ascertain that a value supposed to be the same in both modes really is.

In case the result of these tests would not solve the original question, please feel free to reply.

Best regards.