You could do these manually or what we are trying to do is build a automated system for it and it requires us a lot of testing to make sure we get the numbers exactly right.
Anyway what matters with a lot of these trades is you figure out the direction in the first 5 minutes of market open. Either you look at very extreme divergence between RSI and ADX. Or you can just look if the 2M either have higher highs or lower lowers. Then use RSI to exit the trade.
I dont know if you figured out these things or if your bussiness is just selling training to other noobs because your track record online is not exactly that impressive. 22% win in 1 year almost with 3 last months losing months.
And why these stocks are gapping up so much now you can start to look into pending futures and option orders on each of tickers and see if what happens there. There is some information of course not that open shared on the internet.
Basically, this topic is waste of time because you canât replicate this results on live account. According to my results, 22 percent is still higher results from many investment funds and 3 months stagnation is normal in trading. Looks like you have to gain more knowledge before talking about trading.
1D timeframe, over 500 trades, 22701% win over 1 year and only 2 losing trades.
Try figure out yourself now before you have more stagnation months
Before you go throwing granades into the market and teaching others the same maybe you should try to backtest your own strategies and figure out why you are not consistently making money.
There are 99% bull-sh strategies and ideas. You realize created by people that donât really backtest on historical data do they? Write a book or make a Youtube video for some strategy that maybe works 50/50. I fall into this trap myself then I realize after backtesting on millions of records of historical data (yeah, we have 1B rows in our database on every timeframe) that most strategies and ideas donât work and only a very few rare instances (based on OHLC suprise suprise) actually works??
If you canât program or have access to full inventory of gap-up stocks (need at least 20K in your account) go write a book or something. Or like you say in your bio I decided to take fees for provide information which means you are not profitable trader at all and just wait for the next big great idea.
Bro, come on. There is enough degenerates that trade on live account not having a clue about what they are doing. Winning a few bucks here, losing a few bucks here.
If something works on historical data over 1 year itâs not a problem for us to connect the strategy to a demo account first, run it for a 1 week, see how much margin we are using and try on a live account.
Try doing the same as us, collecting billions of data points and make your own strategies (not copy other people or other people ideas suprise suprise). Then go try it on a live account.
We are not stupid like rest of the internet
Try make a system that works 99.99% of the time. Not 60% of the time. That was our goal all the time
Did you try your strategy on 1 year of historical data? I guess not.
No wonder most trading firms not hiring traders any longer. They are hiring computer engineers and mathematics. Not some gambler
You also start to wonder when you run strategies developed by people such as Linda Raschke and Larry Connors. And itâs not working on historical data. Itâs winning 50/50. Same as flipping a coin. Then you realize every trader asshole made their money selling books and training not having a single clue what they are talking about.
As I suspected, overfitt strategy and try to be smart. I was very active, gave useful information, what I got was rude behavior. Quality information doesnât comes from air, it takes time, testing, draw conclusions, I will not give it for free. You can write what you want but the truth is I am profitable in long term you are not.
Mhm, it doesnât means you can end backtest in 20.10. you use 2500 tickers on demo account. I think we can end this useless conversation, there is no value for me or the community.
I think it is cool strategy he is using, but maybe it is not that effective since we cannot rely much on demo vs live. But as long as he is formulating the strategy that is working out for him, then good for him.
Simple rules:
If price moves under or over MA, then wait for the breakout (meaning the price moves in favorable direction). Itâs important to wait for breakout otherwise you might trade in ranging or choppy markets where the MA goes from green or red every candle. Take exit when RSI reach either 30 or 70.
I am still in process of testing this system with historical data. A few tweaks here and there. Then I will put it into machine learning to see exactly which RSI or MA settings are best combined for each invidiual stock.
What is nice about this system?
According to the backtest over historical data for period of 1 year there was only 2 losing sessions. Using dollar values instead of percentage. A few of my strategies was horrible with invididual stocks losing 400% over 1 year period which this system is not.
Biggest losing session was $250 and biggest winning session was $3.2M dollars.
If system was using with $100K volume on each trade over a year the system would have generated $159M. This is only theorteical because not every position would fill at $100K. To reach Jim Simons wealth of $25B it would take my system 157 years. I guess my system is good but not as great as these guys working at Renaissance Technologies
Improvements:
We break up sessions in days (from 08.30 to 14.30 New York time) because itâs a lot easier to take multiple profits when the trend is strong. The system kind of works like a elevator. When you press a button it picks the elevator further away from first floor and make the first floor rest ready for passengers. However, losses might not exceeed into the next day so must make a more dynamic system to hold losses overnight and trade the next day instead of taking losses at end of the day. This is difficult programming wise to hold multiple trades over longer sessions. If we solve it the simple way and remove sessions all togheter that might cut the losses faster but limited profits over longer trends. Since that would lead to take only 1 trade in long-term trend instead of multiple trades (that the system is currently doing)
You always want to keep one elevator ready at first floor at all times and second elevator to move up and down so the system must be divvied into a two elevator system instead of only 1 elevator system.
In PHP since itâs a single thread language this will become a headache. In Python or Java which is multi-threaded its a lot easier to do, but I donât want to re-program this whole thing over to Python and I can say MQL or Pinescript especially wonât do any better either.
I want to remove session and create a two elevator system. One elevator is ready every trading day another elevator takes trade into the next day. If take profit is not triggered the first trading session I want to take it the second trading session.
Session is great because it lets me take multiple trades during a longer trend but it fucks up my take profit during losing trades that might have a better profit option the next day
One elevator is ready every session and another elevator moves up and down the losing trades. If it can find a better stop loss position
Thanks bro I appricate it. Please give your input. There are bullies all over the place. Thanks for your support.
One elevator is working. Two elevators are even better
First elevator stays open at 08.30AM (New York session). Second elevator moves up and down between different sessions
Letâs say one my trades take a loss in the first session. Instead of closing the elevator and taking it back to first floor I keep this elevator open. My system only works as single elevator system and its still making $159M per year. I am pretty sure if I keep two elevators open I can make additional $50M per year.
Of course the discussion is it worth the extra work for another $50M per year? Or will the second elevator system even make more losses? It must be backtested.
Hi alright guys. Sorry about my outbreaks. No more than 2 shoots per day. Only way to solve alcoholism is to contiune drinking but limit the shoots I guess. No beer. No wine
This algo is made on the open price so it wonât work as a ordinary OHLC algo where you get the price every M30 and then do the calculation. This algo need to get price every M1 then do the calculation every M30 of course a little bit more hassle to look at two graphs at the same time but fully doable.
