charlotte nc
Experience: Advanced
Platform: My Own System
Broker: Optimus
Trading: Emini (ES, YM, NQ, ect.)
Posts: 408 since Jan 2015
Thanks Given: 90
Thanks Received: 1,148
|
Hi,
I submitted a feature request to NT recently and could use some upvotes to help get this implemented.
Here is a link: https://forum.ninjatrader.com/showthread.php?p=540177
Here is the run down. In Market Replay NT SIM engine assumes 0 latency and moves from
1. Intialized
2. Submitted
3. Accepted
4. Working
In 0 milliseconds. So this execution assumption would be appropriate if you were plugged directly into the exchange. Most retail traders are no where near this, so this assumption renders most market replay runs overly optimistic.
By Contrast in live SIM, this exact sequence takes around 100 to 150 milliseconds. Now if you are averaging around this in your live trading, then you will be happy because this is running roughly the same speed, so this is a fair proxy for it. But most people with a VPS / proximity trading can get between 1 Millisecond to 25 milliseconds in this execution chain. So this understates that speed by a magnitude of 5x, 10,x 20x.
So my proposal is to have the execution speed be user defined. For my particular use case neither of these assumptions are in line with my live execution speeds. I am far closer to the market replay speed then the SIM speed, but I can't work with a 0 latency assumption and feel like I wasn't lying to myself a little.
So if you would find this feature request useful, please up vote it. I think that is how these things work. The features that get up votes get built. I believe user defined execution speeds would make testing considerably more accurate.
Thanks,
Ian
In the analytical world there is no such thing as art, there is only the science you know and the science you don't know. Characterizing the science you don't know as "art" is a fools game. |
|