somefool
New member
I'm assuming one or more of my points below is not 100% accurate so I welcome anyone to correct me. I apologize in advance for writing in frustration. Please take that into consideration.
New to this forum and thinkScript but not new to TOS or trading. Have had some success with discretionary trading but am ready to take the next step and translate my methodologies into thinkScript Strategies so I can back test and forward test my ideas and use these signals to enter real trades in live trading. Eventually I would like to move on to fully automated trading. I've taken some programming classes (mainly C++ and C#). I'm not programmer but can read and cobble together some code and do a few things.
After weeks of studying thinkScript I'm about to give up and try a different platform like TradeStation or NinjaTrader. I am frustrated that thinkScript's AddOrder function only executes trades on Open of the next bar instead of now if my condition is met (while the current bar is still being drawn like in real trading). After years of discretionary trading I'm painfully aware that bad things can happen between now and open of the next bar . Entering a trade on open of the next bar is not illustrative of real trading in my experience. You can do that if you want, but many traders have no desire to wait that long. Because of this apparent AddOrder limitation, the backtesting reports thinkScript generates are inaccurate imo, rendering them essentially useless. I cannot trust the reports, so I cannot truly test my Strategies.
While it is possible to code a Study (or Strategy) indicator and follow its alerts (which do appear to be immediate) and use those signals to manually place real trades during live trading, it is not possible (to my knowledge) to accurately back test or forward test a Strategy which kinda deletes the whole point of thinkScript Strategies. I actually saw another person ask the same question this week somewhere else on the net and the answer was to tweak the reports in thinkScript so they model results based on orders being executed now, instead of based on orders being executed on open of the next bar. Hmm. I guess that might work but haven't tested it yet. Curious if anyone has had any luck doing a tweak like that or if I am not correct about this limitation of AddOrder?
Not to run my point into the ground, but just for clarity TradeStation, for example, has two modes;
On Bar Close Order Generation (Default)
Orders are generated at the close of this bar to be filled on the next bar
Intra-Bar Order Generation
Orders are generated multiple times on this bar to be filled this bar
Thank you
New to this forum and thinkScript but not new to TOS or trading. Have had some success with discretionary trading but am ready to take the next step and translate my methodologies into thinkScript Strategies so I can back test and forward test my ideas and use these signals to enter real trades in live trading. Eventually I would like to move on to fully automated trading. I've taken some programming classes (mainly C++ and C#). I'm not programmer but can read and cobble together some code and do a few things.
After weeks of studying thinkScript I'm about to give up and try a different platform like TradeStation or NinjaTrader. I am frustrated that thinkScript's AddOrder function only executes trades on Open of the next bar instead of now if my condition is met (while the current bar is still being drawn like in real trading). After years of discretionary trading I'm painfully aware that bad things can happen between now and open of the next bar . Entering a trade on open of the next bar is not illustrative of real trading in my experience. You can do that if you want, but many traders have no desire to wait that long. Because of this apparent AddOrder limitation, the backtesting reports thinkScript generates are inaccurate imo, rendering them essentially useless. I cannot trust the reports, so I cannot truly test my Strategies.
While it is possible to code a Study (or Strategy) indicator and follow its alerts (which do appear to be immediate) and use those signals to manually place real trades during live trading, it is not possible (to my knowledge) to accurately back test or forward test a Strategy which kinda deletes the whole point of thinkScript Strategies. I actually saw another person ask the same question this week somewhere else on the net and the answer was to tweak the reports in thinkScript so they model results based on orders being executed now, instead of based on orders being executed on open of the next bar. Hmm. I guess that might work but haven't tested it yet. Curious if anyone has had any luck doing a tweak like that or if I am not correct about this limitation of AddOrder?
Not to run my point into the ground, but just for clarity TradeStation, for example, has two modes;
On Bar Close Order Generation (Default)
Orders are generated at the close of this bar to be filled on the next bar
Intra-Bar Order Generation
Orders are generated multiple times on this bar to be filled this bar
Thank you
Last edited: