Ninjatrader realistic fills?

[nourozi]

Yes, that's what I would expect, the speed should not be affecting the sequencing of the events seen in the streams.

Hi again,

I have just compared the same weeks trading results with an automated strategy using both Market Replay and Live Simulation. I ran the exact same setting on Live Simulation as I did with Market Replay but the difference was quite large.

Basically I was getting filled much more easily/faster with Market Replay than I did with Live Simulation.

I ran the same strategy, over the same week, using the same settings with Market Replay and Live Simulation and got largely different results.

I thought the results were meant to be similar?

Why would there be such a big difference?

[pstrusi]

Interesting situation cause it shouldn't, according to NT market replay logic procedures. I got a questions:

- Does this situation occur in others else instruments or markets?

- According to Bertrand, whether you use market replay or live simulation, the size of your orders should not have any influence in market dynamics, but now I wonder if the internal NT engine simulation could change cause of it. Were your orders's size small ?

[nourozi]

Interesting situation cause it shouldn't, according to NT market replay logic procedures. I got a questions:

- Does this situation occur in others else instruments or markets?

- According to Bertrand, whether you use market replay or live simulation, the size of your orders should not have any influence in market dynamics, but now I wonder if the internal NT engine simulation could change cause of it. Were your orders's size small ?

I did live simulation across a couple markets and both had large differences between live sim and market replay. I am also only using 1 contract.

There are a couple factors I can think of but wouldnt think they would have that much impact on the results:

Data feed of live sim is Rithmic compared to the Zenfire feed for Market Replay.

The simulator delay settings do not apply in Market Replay but they do in live simulation. I am talking about the Delay Comm and Delay Exchange, also the non simulated delay of data from the exchange to your PC. These 3 delays are not present in Market Replay.

The only other factor I can think of is that maybe the live simulator algorithm engine is different to the Market Replay engine?

[pstrusi]

Well, it seems there must be cause more complicated reasons then; looking forward to know appreciations of NT support team regarding your report.

[NinjaTrader_Bertrand]

When compaing you would need to keep in mind that any delays set are not factored in for replay.

[nourozi]

When compaing you would need to keep in mind that any delays set are not factored in for replay.

So you are saying the fill algorithm is the same for Market Replay and Live Sim apart from the delay being factored in?

It would be great for Ninjatrader 8 to have the fill algorithm for Market Replay to be the same as Live Simulation.

[NinjaTrader_Bertrand]

Correct, the underlying 'engine' is the same.

[nourozi]

Correct, the underlying 'engine' is the same.

Interesting, I wouldn't think that < 200ms difference would make such a difference between the results. Especially considering I am using resting limit orders that are sitting at the exchange for over 30mins on average.

200ms out of 30 mins is nothing...