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)
Thanks, I responded on PM. Tried to remind artemiso to send me him email, because I can only send 1 pm in an hour LOL This forum restrictions are terrible....
Its all very interesting but I think you over complicate stuff. Day like today when market moves no were is best for scalping. For manual trading keep simple your brain can process only so much info. Ex wait for your Level when you see absorption and orders were added to bid and subtracted from ask put a limit order . In trending market you will get 8 tiks min day like today 3 tiks .
Interesting on how do you classify "a day like this"... If it was as easy as you say "absorption..." then "orders" added to bid and subtracted from ask..put limit order...and you will have as many wins as you show on todays screenshot... then you should be making a lot of money most of the time?
Do you mind sharing several months of stats of this strategy that you describing?
Something tells me that this is very discretionary and not consistent. I might be wrong... If it's rules based and consistent why don't you automate it via Rithmic API and have a bot doing with 10ms second speed? If it's something serious and you are interested in automation..I can do it for you.
Enozer factor I use SPEED . On picture you can see levels were broken with 3 times ev speed (stop hunt). Its a signal to take a close look at order flow.
Here what I was talking about . Price at support , orders adding on a bid side , plus a 100 lot buyer (dimond) . Went market to get this one , easy 5 ticks .
I have used gradient boosting as part of my day job before, but I haven't really given it a go on futures data before.
I have no doubts that with a decent training data set, I could get a small bump in performance. But in practice, I don't think that a guy at my level (retail / lone wolf) could run R-studio, or Python in real time with any benefit because those would make speed an issue.
The best case would be to run some ML on the data, and then figure out what the actual insight was that held the edge and then reverse engineer that into the actual code logic. But trying to derive the logic from a ML program is kind of a trick. So I haven't really pursued this route.
If you have any ideas, or advice I would be all ears.
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.
Hello,
First of all thank you for very interesting thread.
Could you, please, clarify how you calculate " [Bid/Ask] SS L2" columns (file "ES 2-25-19 Summary Statistics.xlsx" )?
At row 14 where Ask side win "SS Ask Starting V"[14]==" Ask1"[14]==113.
But I cant understand how "Ask SS L2"[14]==218 got.
Thanks for taking an interest. The level 1 and level 2 data feed are not perfectly in sync with the program that I used (NinjaTrader). So there are some gaps and data mismatches here and there. If you look at column Z row 16 you will see the 218 on the level 2 feed. On row 14 this showed up as 214, so this is likely what you would expect to be used, but the sequencing in my extraction program with the data lagged picked up the value of when it moved up to 218. In most cases the difference will be negotiable on these, but I see where your question comes from.
Ultimately what this metric is doing is showing the best bid / best ask volume for each price along with the corresponding best bid / best ask for the top level in the level 2 feed. Getting level 1 and level 2 to match up is a bit of a trick. Because with Ninjatrader the level 1 and level 2 feeds come from different event handlers and process at different times. So I have to create a lot of different temp variables to hold everything to get these to two feeds to sync. Sometimes no matter what I do, I am ahead or behind by 1 row, but the difference is just going to be what changes occurred in the level 2 feed with just a few ads or cancels here and there.
I have never seen anyone else share this level of analysis on something as massive as the full book on the ES. There are likely better tools / ways to go about this, but in terms of free tools / dirty analysis, this is likely about as good as it gets.
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.