Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Personally, I love the data science and intellectual challenge of high frequency data but the older I get the more I hold common stock and do a lot of sitting on my ass doing nothing as Charlie Munger would say.
I am sure some people make great money trading order flow but I am even more sure many are delusional and/or just making things up.
I mean the idea there are all these misprisings and alpha waiting for you at home on the order book is just not rational.
Following up on my last post: Here is the actual analysis and raw data.
I won't provide a long winded explanation of the technical aspects of the analysis. Everything is in excel and you can easily trace all of the formulas to see how I did everything. I think I used a fairly reasonable approach, and even though I didn't do a ton of QC on this casual analysis I think it is directionally fairly accurate.
Thanks for reading and taking an interest!
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.
Very impressive info.
Did you happen to try take advantage of this yourself?
If I did not misunderstand something, you mentioned this type of things are not accessible for retail traders?
Do you think having Rithmic or CQG access and your own code would not be able to take advantage of some of the patterning that perhaps exist there?
Thanks
One of the reasons I am also asking this is because I am working on one trading system of mine that involves support/resistance price levels and wondering if it's possible to predict a break through with a higher probability of success by using some method of analyzing the DOM activity at that level?
I am assuming that I would write a code for this of course.
Thanks
You might have a shot with rithmic's diamond API, but that is about the only one on the market (in terms of retail-ish).
The edge is solid, you can test it over any day, any time of day, etc, and it will generally produce around 55% / 45% in favor of accurately predicting the next major price level break.
If you had an seat license, exchange / broker discounts and were getting 100- 200 trades a day, you would still need around 60% / 40% to make money. This edge alone is not enough. But with around 5% more cowbell you can mathematically break even.
I've been too swamped with work, and other pet projects to spend much time testing things like this. I did have a go with Ninjatrader a little while back with spectacularly shitty results. Their platform crashed during live trading, they later did a major revision to the edition to fix the bug, but ultimately they weren't fast enough to be able to capture this edge. Even with a VPS with a decent co-lo, the issue was that their data feed event sequence has a built in lag.
I am sure this is the case with a lot of retail tools, but prioritize a thread that most traders will use, and the more granular data feed gets a lower priority.
When I pick this back eventually, I might give rithmic a go, and possibly look at something like treasuries. I think there might be a higher edge with books that are thinker, and mathematically the gap to break even may be far less vs. the ES.
Best of luck out there.
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.
Thank you for the response. Do you mind sending me a pm. I can't as the forum says that I am too new.
I could share some working things I got done so far, plus I am working on this stuff full time and could potentially help out with some testing.
Hi quants00, I am always happy to share and help if possible. If you have any specific questions, let me know and I will be glad to give you some feedback.
But i will always defer to @artemiso on this topic. He is a legit pro whereas I am a hobbyist. Because I don't do this very seriously at this point, I have been willing to share what little secret sauce I have found. I am a data scientist by trade, and this world definitely fascinates me, but I haven't had much time to put into it lately unfortunately.
If anyone has any interesting thoughts on this thread, this topic, or would like to add to the discussion I think there are some cool things here, but most retail tools will have no chance of working at this granular of a level.
I''ll definitely PM you though to follow up.
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.