Stunning Strategy Analyzer Results. Is this realistic?!

[traderzoso]

Hello,

I've been running test in the Strategy Analyzer and the results have me falling out of my chair! My question IS, how close to reality are these results? What percentage to reality would these results actually be?

I used a professionally coded automated algorithm for the test.

I ran the test for this month of August 2019 only. Total net profits (one contract) $179,562.00

This massive profit, seems ridiculously too good to be true. What do you think?


See screenshots attached.

Thanks,
Ern

[chartish]

Not even close to being realistic.

[traderzoso]

Ok, and now the 100K question....why?

[chartish]

I can't critique your specific strategy, but these results look pretty preposterous.

[traderzoso]

I hear ya, but it is a very well done (expensive) algo. I just put it on an hour ago in NQ, ES and RTY and took a max profit of $730.00 combined, one contract each. So we shall see. I want ot get some feedback from Ninja folks and thanks for your comments.

[chartish]

I don't think this is the place for this discussion, but good luck.

[traderzoso]

I'm just wondering how it would be possible for the analyzer to produce "pie in the sky": results, and or how close to reality it is. As even if its 30% or so close to the results I'd be happy. I want to be able to understand how that would take place, and specifically WHY.

[traderzoso]

Where would be the place for such a discussion?

[chartish]

Elitetrader.com or futures.io is where I would post.

[traderzoso]

I see, well I want the folks that are directly involved with the development of ninja platform software (Strategy Analyzer) is why I'm posting here. I'm looking for technical feedback more than 'am I gonna make boatloads of money' kinda thing.

[chartish]

One thing I noted is that you run it without commissions. I would always backtest with commissions. Your simulation took a lot of trades, so commissions is important.

[traderzoso]

Yes, good point, I normally do. Thanks

[SuperDriveGuy]

Yes, good point, I normally do. Thanks

You have 1859 trades, even if you account for commissions(say $4 per trade) = 1859x4=7436, so net of commission too its a good strategy.

The only things that comes to mind are fills and slippage. The fills may not be possible in the real world, so you MAY lose a tick on entry and exiting the trade. Volatility may also lead to slippage in some cases, during news etc

I would also suggest to run in it replay mode and rule out outliers and if it still is profitable, hey you found the SECRET to trading success!

[RJay]

Hello,

I've been running test in the Strategy Analyzer and the results have me falling out of my chair! My question IS, how close to reality are these results? What percentage to reality would these results actually be?

I used a professionally coded automated algorithm for the test.

I ran the test for this month of August 2019 only. Total net profits (one contract) $179,562.00

This massive profit, seems ridiculously too good to be true. What do you think?


See screenshots attached.

Thanks,
Ern

Traderzoso, It appears your code is entering and exiting your trades at the wrong end of the bar in your back-testing. Special coding is needed to compensate for this. Replay and Real time trading are probably OK.

[NinjaTrader_Kate]

Hello traderzoso,

Thank you for your post.

Backtesting uses OHLC values from the bar to simulate order fills and actual market movement is not used. Renko Bars also change the open of the bar, which cannot be simulated with historical data. Therefore, it is generally not advised to consider Renko backtest results as being reliable. You could use Market Replay data in the Playback Connection and run the strategy there so it is simulated as if it it were running with realtime data, which will show more accurate results, RJay is right about that.

Please let us know if we may be of further assistance to you.