Stop Loss work differently in Backtest and Real Time simulator

[stefy]

Hi

I use ATR as stop loss. In the case of GBPCHF, 1 ATR(14) amounts to about 28 pips in a normal day.
When I backtest, the Stop Loss kicks off sometimes, but when I use the Real Time simulator to paper trade, I almost always get stopped out by my stop loss. If the day after I run the backtest, often I don't.
I guess it's due to the fact that the Real Time simulator uses the Bid/Ask prices (which is already 10 pips with Gain!), so a small movement of 15 pips + spread can already kick in the stop loss.
Am I correct or there is another plausible reason for this to happen?

If this is the reason, how can I amend it practically? When I backtest the model it's very good, but when I use the simulator many times the positions are stopped out before they start to be profitable.

Thank you

[NinjaTrader_Ray]

You could try a wider stop setting when executing in real-time?

[stefy]

I actually thought about this... but I wanted to be sure this happens because the Real Time simulator uses Bid/Ask prices whilst the Backtester uses simply Last price data.

[NinjaTrader_Ray]

Yes, simulator used bid/ask for triggering orders while backtest uses the bar information which with GAIN is driven by the bid price.