I'm backtesting a system using markets orders only.
I compared the results trade per trade of a strategy running with no slippage and a strategy running with 1 tick of slippage.
Sometimes for the same trade, the P&L difference is only 1 tick instead of being 2 (entry and exit).
Looking at the charts, I noticed that this is coming from NT not taking into account 1 tick of slippage for a buy if the high of the first bar after the signal = the opening price of that same bar (low of the next bar = opening price for a sell).
Please see attached charts. The 2 charts on the left assume 1 tick slippage and the 2 charts on the right assume no slippage.
The 2 charts on the top are the same trade with different assumed slippage, the 2 charts on the bottom are the same trade with different assumed slippage.
In the top left red square NT assumes that because 3151 was the high of the bar, it couldn't have any slippage. However 3151 could have been the bid at the opening and my market order would have printed 3152 in real conditions.
Same for the closing trade in the yellow square on the bottom left.
I don't know if this is a known issue, but it needs to be corrected.
Hope it makes sense

Comment