Hi, I decided to try out Tradestation's backtesting engine. I'm playing around with a simple strategy which enters on 60m bars and which makes use of the built-in stop function "setdollartrailing". I've realized that the only way to get correct simulation of stops is to use look-inside-bar with 1 tick but then you can't backtest very far back. Not using look-inside-bar gives rise to huge differences in some cases. So the only way to "properly" backtest with stops far back in time is to stop out at the next bar? For example by using 1m bars as data1 and 60m bars as data2 and trigger stops as data1 breaches the stop level(?)

For those of you experienced in Tradestation, have I understood this correctly? Exactly how do the built-in stops work in simulation when not using the look-inside-bar testing?

Thanks

Don't use setdollartrailing or setpercenttrailing in backtest, ever. That is my rule.

I have no idea if they work in real time.

Thanks for your reply.
I have monitored those functions in sim-mode and they seem to work in real time but the backtest results can be way off.
Say I watch the strategy executing live throughout the day and take a screenshot of todays trades from the "Strategy performance report" at the end of the day. Next day, I backtest the same script and look at yesterdays trades in the "Strategy performance report" and compare to my screenshot from the day before. The backtested exits, executed via "setdollartrailing", can differ a lot from what I observed live the day before. I have instead coded trailing stops directly on 1m bars while the entries are done on 60m bars.

This is a known longstanding "feature" (issue) with setdollartrailing and setpercenttrailing. I don't allow traders I work with to use them at all.

Kevin, What are the alternatives to setpercenttrailing?

Is this the only option? Tradestation SetPercentTrailing Replacement (https://community.tradestation.com/Discussions/Topic_Archive.aspx?Topic_ID=8714)