strategies tab

[Dobbes]

I have a question about the strategies tab off the control panel. What exactly is the difference between the real time and real time and historical strategy performance?

More specifically, what historical data is being used? It doesn't seem to be the backtest data, since the number of trades executed does not match with the backtest. Is the historical data just from a log of the strategies stored trades? Is the realtime data just from today, or is it from the last time the strategy was started?

If I stop a strategy, and then restart it, will I need to view its performance under historical and real time?

[NinjaTrader_JoshP]

Real-time - Selecting this menu will generate performance data for your real-time trades only (since the strategy started running) and will exclude historical trades. If your strategy held a virtual position (calculated against historical data) upon starting, a virtual execution representing the average price of this position will be injected into the real-time results to ensure that a trade pair can be created with the executions resulting from the closing of this position.


Historical & Real-time - Selecting this menu will generate performance data for both historical and real-time trade data.

Backtest results can be different due to the following reasons: http://www.ninjatrader-support.com/H...sBacktest.html

[Dobbes]

I read the manual too. Doesn't answer, WHAT exactly is being used as historical data since it most certainly isn't the backtest data.

[NinjaTrader_JoshP]

Not sure why you are seeing a difference. If you run a backtest on the same period it will be exactly the same.

[Dobbes]

So the historical data should be comparable to the backtest, but for some inexplicable reason it isn't. I guess that answers my original question, but now I have no idea why the two don't match up. Let me see if I'm understanding correctly: In theory the historical + real time strategy performance should use the real time data as a continuation of the backtest, so the #of trades should be backtest + real time. Is that correct? Or am I missing something? Because the #of trades != backtest + real time.

[NinjaTrader_Ray]

It would be the same if the underlying data set is identical meaning, the start date is the same.