It's not just NinjaTrader. You can see issues from the datafeed you use as well.
For instance, you run a new strategy in sim on real time data, and then you run it again afterwards on exactly the same time period, and you can easily get differences. I've seen it be so bad that my strategy in real time took a short trade but in backtesting on the same data, it went long! Sounds nuts until you experience it yourself and investigate exactly what happened - a tick different here, a timestamp different there - and suddenly the indicators are different and then the trades are different.
And that wasn't on NinjaTrader, that was a home-grown Java app I wrote.
But you are on the right track. The longer the time-frame you use, the less you run into problems like that.
The whole back-testing thing is a learning curve, but I guess what you are asking is whether it's worth it. You'll have to work that out for yourself though.