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)
I can not believe the statements of NOT making money some posters here are playing out.
In contrary I have 2 journals here on fio - one system is semi-automated - means the trades are alerted by SMS before and then I start the trade upon such an alert manually. Nearly all trades had been announced here in the journal before being taken. It is a very simple pattern recognition system which gave over the years a good bunch of positive points.
Last year just 7 trades with a mean of 56 points (x25 EUR/point and contract) which is consistent and obviously destined for a fully automated system. Of course it is only for high volatility days (with exclusive day trading). Only the year 2017 gave not a single trade
Here the glimpse on the results 2018 from the journal:
All the rules are open for free in the first post of that same thread. As long that thread is not closed or post 1 changed
As we can see - consistency and accuracy in very different periods are really possible on a given instrument.
Plus the program is not 30k lines - but in reality less than 100.
Just a glimpse on great pattern recognition within charts - that can be found by thousands.
Most often after long testing sessions i put one idea more at the "one million systems failed" bin. I think that reliable patterns are very difficult to find. My common story ends with very good months results but unable to filter out the loosing streaks.
I dont have such Things like timeframes or something. I use 1 Volume data Series for calculating. All calculation of my algo are based on orderflow and Limit orders. So it dont depends on different market situation.
Lately I am focused on DOM trading , watching the ebb and flow, hitting when coinciding markets align. Automation of this approach would be very challenging (if not impossible) as the process would need to establish " a sense of feeling" which is the cause at that moment to enter as well as exit - its more of an art than discrete science. Its definitely not apparently based on any signally identifiable, normalized, gauged measure or number. Yet, it's not complex either. Is it going to go, Yes, get in. Is it done, Yes, get out. Are you wrong, get out. Context is king as well as confidence and decisions, which are based on everything in/at the moment.
I have found curve fitting to be useless and have come to believe there is no equation to correctly define the market - too many unbounded forcing factors and constants for any math equation I have attempted to derive. As stated in other posts it's software so with money and time it can be solved. I think the solution is rooted in observation of action and reaction.
automation doesnt work most of the time.As the profit we books depends on the entry we make. Thes successful entry can be done watching the market as market is a dynamic beast.
AND automated systems can be dynamic and really need to be . I was a discretionary trader for over a decade and a decent one at that but my discretionary results cannot come close to competing against my mechanical systems . AS discussed by previous posters its the skillset required to produce robust logical systems that stand the test of time . Its easier to become a doctor/lawyer . MAxover , macd . rsi , stoch, fibs ( insert any generic POS ) aint going to cut it . If you cant do it you are the weak link , end of story ... YOU NEED TO INNOVATE , be unique , find uncommon edges . DO what everyone else does and expect above average returns is delusional . What you measure and define to produce a decision tree that produces signals is what matters . It isn't easy and i wouldn't have it any other way , its what makes it worthwhile. When you say it CANT be done all you are saying is YOU CANT do it , don't impress your limitations upon others . This takes years not 3 weeks of curve fitting , needs a special way of thinking , of course not many can do it , it's the way it is . Most people fail at trading and auto is the pinnacle so the failure rate is exponentially higher . Most here should probably go do something else , no offence intended . Reality is a bitch ..
Edit ... If you think every builtin indicator on every platform has not been MLed , optimized , curve fitted and manipulated to death think again . Its a horse beat to death 1000's of times . The answer lies outside this . Time to build a new " Horse "