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)
This one opened my eyes. I almost subscribed for Daytradr and Bookmap...
For futures daytrader at micro levels it is not worth it. I went through No bulshit trading course but it felt outdated.
I will stick with free Tradestation tools and Tradingview.
Also, I usualy trade after hours globex and action is slow at micros MES.
Can you help answer these questions from other members on NexusFi?
Hi, sorry I'm not being as responsible as I should be. So I had a full dedicated reasonably fast Quad Xeon in New York City, but the network latency was limited (who'd uh thunk it?) by the speed of light, and routers.
So I was looking for something, and found a provider, not in Aurora, but in Chicago with super-fast dedicated servers. Currently I use a Dedicated 6 Core Xeon 32gb system (I don't need that much memory currentlly) which is seemingly super-fast, from a provider I could recommend.
This forum seems manic about threatening to ban users for anything remotely commercial... so I'll just disclaim any connection to the provider, and recommend HiVelocity.net. I am running what I consider to be a monster system with the following specs:
Intel(R) Xeon(R) E-2236 CPU @ 3.40 Ghz, 6 Cores, 6 logical processors (since I requested they switch off hyperthreading) with 32gb DDR4 memory.
But its actual CPU speed is running a variable clock 4.60+ Ghz (performance in Task Manager)
In my view, this system screams, and it is <1 msec from Google's DNS 8.8.8.8 by ping, hosted in a super fast network. Way fast enough for me ! and I am a speed freak of sorts...
BUT NEVER FORGET THAT WINNING IN TRADING IS NOT NECESSARILY AT ALL DEPENDENT UPON RAW SPEED AND POWER. YOU NEED TO PREDICT THE MARKET'S FUTURE PRICE. That's a whole 'nuther kettle of fish...
My software is insanely multi-threaded, and this thing screams reliability and performance. Ninja Trader 8 is (for an off the shelf retail platform) just screamingly fast, when used correctly.
Hosting provider support is solid as a rock, which is something you have to consider.
My cost monthly is $200, and it's worth it at twice the price.
So consider that; and if you're concerned about the monthly cost; then you may not be ready to go all in ..... For my own custom software in C# hosted within Ninja Trader 8, I'll outstrip even a heavy VPS (shared type) system in no time.
A machine with that power, and Reliability, you could share with a trading buddy.
It's self-managed; and I assume you know what that means !!!! I have personally tuned it more or less like a "gaming machine" since Trading is a GAME... LOL just kidding; but the latency requirements are similar.
In my case, I am sitting out in Idaho; but I don't tell the Chicago system to "Hey, Buy now... right now!" since the latency of display in the Nasdaq 100 Futures market makes that impossible, so "manual" Order Entry is really not going to be very successful. My fully custom Strategy software uses a "fire and forget" custom Order Entry system, so I can set a dozen trigger conditions, "arm the system" and let 'er rip, so the Server hosted software decides when to strike, and manages up to dozens of simultaneous orders, targets, etc...... I use RDP access, and that spans TWO gaming displays. The pic is only of the rightmost display I typically use for NQ/MNQ mixed trading.
So I'm in pretty deep here; but the provider I'm recommending has truly been a great service for my system... Your mileage may vary, as they used to say... I'll post a pic of my order entry client which also does all that analysis; this is NOT for sale, so no advertising here...
[EDIT] I forgot to say, relevant to this thread, that the Analysis also does Order Flow, which the PURPLE line would represent; but it's a very particular type of "Big Lot" Integration of Trade Flow; and, oddly enough, a fairly simple algo like that will nail Tops and Bottoms; but there's a lot more that my system does; arguably TOO MUCH, as you can see...
[EDIT] we've settled down into the morning session here; and at market open NQ (through the Rithmic market feed) generates in excess of 1000 events per second; and I take data from NQ/MNQ simultaneously. Here are data rates, like I said, pretty slow now, as we're 2 hours into this session:
MarketDepth Updates minute 22 Primary: 34389 Secondary: 40511 Ignored: 10825
MarketDepth Updates minute 23 Primary: 26928 Secondary: 31472 Ignored: 8365
MarketDepth Updates minute 24 Primary: 19454 Secondary: 21746 Ignored: 5839
MarketDepth Updates minute 25 Primary: 16051 Secondary: 18549 Ignored: 4961
MarketDepth Updates minute 26 Primary: 24360 Secondary: 28140 Ignored: 7092
MarketDepth Updates minute 27 Primary: 19145 Secondary: 21755 Ignored: 5446
MarketDepth Updates minute 28 Primary: 14684 Secondary: 16516 Ignored: 3999
MarketDepth Updates minute 29 Primary: 20505 Secondary: 24195 Ignored: 5835
MarketDepth Updates minute 30 Primary: 19542 Secondary: 22958 Ignored: 5814
MarketDepth Updates minute 31 Primary: 21908 Secondary: 26492 Ignored: 6578
MarketDepth Updates minute 32 Primary: 22686 Secondary: 28614 Ignored: 7092
MarketDepth Updates minute 33 Primary: 19728 Secondary: 24372 Ignored: 5868
MarketDepth Updates minute 34 Primary: 15943 Secondary: 20057 Ignored: 4746
MarketDepth Updates minute 35 Primary: 18052 Secondary: 21848 Ignored: 5550
This stuff is so fast that I have to Queue it for later processing, and return from the
OnMarketDepth callbacks, etc in Ninja... So a fast system is highly desirable at times...
Consider the typical trader; I would very much like to be able to help that person, but we immediately come up against the need for technology to provide an "edge".
There's a huge market for Indicators, many of them focus on "Order Flow" and related concepts, claiming to give you an edge.
I honestly don't have time to spend doing something like this, but maybe another Thread could be started for NinjaTrader 8 users, who are of medium competence in creating an Indicator, for whom "code snippets" and design ideas could be shown.
It would then be up to each trader; or some hacker who volunteers, to build our Order Flow Indicator, with the aim of delivering 50 - 100 Top and Bottom signals for Scalpers.
I could throw in some code snippets; but mostly would try to contribute design concepts; and a rationale for why, and then what we're trying to measure as our "signal".
Such a thing quickly outstrips what you'd want to do with the built-in NinjaScript Editor; but the various classes can be "inlined" and done that way, so long as the codebase text doesn't get too huge; otherwise, it's Visual Studio and DLL generation, which has a few issues, but is much more scalable.....
C# coding would be a must; and it could be "fun"... LOL
HOW WOULD SOMETHING LIKE THAT BE RECEIVED BY THE COMMUNITY?
Trading: CME Futures & US Treasury Bonds Futures (week) and Bitcoin (week-end)
Posts: 88 since Feb 2021
Thanks Given: 95
Thanks Received: 65
Look at what tools institutions are using and what strategies they are using and try to implement them.
You might never know their strategies, but often I see a DOM, T&S or FootPrint on their screens in the trading floors.
Thanks for your comment. However, what makes you think
that institutions have tools which will help individual
traders?
What makes you think they know what they're doing; as
opposed to just having a huge amount of financial strength?
Uncritical comments like that are common with Traders;
no offense to you personally...
Only focussed Research and Development, which is a fancy
term for "trial and error" can begin to approach a solution
that will work for a Trader; that's "my story and I'm
sticking to it" as they say...
Generic tools are just going to produce results which are
unlikely to Win; or even Survive.
There's all sorts of published research studies about market microstructure that help explain how both the market making and execution for clients sides of large institutional businesses work. Studies that show edges based on mathematical properties that are the result of the rules a limit order book operates on. Some of these edges are things that retail traders can use or learn from.
For instance, newer research is based around watching the order flow with a neural network, and these studies appear to show some efficacy. Using a neural network is in a way duplicating the way a discretionary trader would trade using the DOM. The DOM is the most effective and efficient method to show the trader everything about the order flow. It also facilitates being able to watch multiple instruments at the same time. There are things I can see on the DOM that are quite clear to me, but I cannot get my various ways of graphing out orders to show these signals. Then on top of all of that it's the best way to quickly manage placing and canceling orders.
Having worked for a large bank believe me when I say that they know what they're doing. They're not just winging it. There's a very intentional and scientific way that they go about discovering edge to facilitate their trading. Their edge is more than just having a stop so far away that it'll almost never get touched. They are able to compile massive amounts of information that have real measurable edge, and execute on that information efficiently in a way no retail trader can. Granted their DOM has execution features for icebergs, auto stacking/pulling, etc that our DOM's don't have. However, the fact that despite all the tools at their disposal the DOM is still at the center of their arsenal speaks strongly to how effective of a tool it is.
Seth, that's very interesting. However, you can speculate how
"quants" and math geniuses facilitate an institution's success;
but my point is that individual traders have different requirements
and constraints; not the least of which is that they generally
do not have "deep pockets".
I've no doubt Institutions are not 'winging it" but what they are
looking at is at a "scale" which in inappropriate for the needs of
the individual Trader.
Everything you say is undoubtedly true... But just assuming that
an Institution has 'the answers" and that individual Traders
who seek an edge should look to "copying" or being "inspired by"
what Institutions (and their "quants") are doing; doesn't
to me seem a valid piece of advice, relevant to Indicators we
can design for our Individual Trader needs.
