Limitations on backtesting, is this a problem with MT4?

I am using NinjaTrader and have written a few systems for automated trading however the limitations to NinjaTrader when backtesting are that if you call the GetCurrentAsk() or GetCurrentBid() it gives you the closed price of the evaluated bar meaning I cannot use my custom coded trailing stops and do not get the correct buyin as my process is a very short timeframe scalping method thus it appears I am limited to the open,close,high,low data for backtesting and not actual tick data, is this the same for MT4?

backtesting on mt4 is absolutely meaningless for timeframes below 1H.. it only makes sense for forward testing, but with those spreads the results would be non significant (im assuming you scalp somewhere else)

Gotcha, this is unfortunate oh well, thanks for the response and yes I scalp somewhere else

I've done the following simple experiment many times:

1. Forward trade EAs on desired timeframes. (I've gone as low as M1, but mostly M15+).

2. After awhile (a week or more), backtest over the exact same time-period.

3. Compare the results.

The gosh darn truth is that the results compare quite well, with a discrepency well below 5%, usually spot-on, even without taking the extra precaution of banking a tick-repository to better simulate the immediate spread. (You can find modules that will do this, if you want nearly perfect simulations ... check MetaQuote's forum for lots of add-ons for tick-simulation purists.)

Just my experience. Try it yourself and see. You might be surprised how well the backtester performs.

Problem is most of my stuff is on 1tick or 1sec charts, max 30sec charts

Just use the backtester in every-tick mode. Without the add-ons, the ticks are simulated. But follow MetaQuote's forum instructions to make the backtester use real ticks. It will then backtest tick-by-tick w/ real tick action. Do the comparison experiment for a day or week (or however long you want), and compare the results for yourself. My guess is that it will be accurate within 5%-10%, if not better.

Otherwise, you can try TradeStation, which features a true tick-by-tick w/ variable spread backtester. You can also use a genetic algorithm plug-in to find optimal results faster.

By the way, tick-charts or sub-M1 charts are possible to generate in MT4 w/ the proper indicator. Again, search MetaQuote's forum ... there are lots of that over there.

Thanks for the input, I will check it out