So this is my best setup for manual trading systems, see screenshot
I call it my RR (Rinse&Repeat) system where i want to see at least 5/6 stars to buy:
- Price in an uptrend (this is a must)
- Price close to trend-channel support
- Price close to continuation pattern support (this is a must) (meaning i want to buy at support OR at breakout -> re-test (resistance has become support))
- Volume profile is not showing distribution (this is a must)
- We are in wave 3 or 5 in elliot wave theory
- Support in trend channel and continuation pattern matches fibbonacci levels
Bonus point: Get a purple or better yet, a red triangle from my custom made indicator, mainly macd, rsi, and someother indicators mashed into a simple “buy the dip signal” indicator.
Stop loss is where you would wanna set it + 5% extra because ive lost so many trades on just being tapped out before the rise, so im willing to risk more.. Ive tracked my winrate and i see that trades with 1-3 stars are looking at roughly 20% winrate, 4-5 stars are at 50% and 5-6 stars are about 70% so far.. Not every trade can be a winner but as long as i find stocks that look like these trades, im pretty sure i can make a profit in the long run.
Target is usually around 15-30% depending on the type of stock, how fast it moves etc. i very rarely manage to “hold to the top” as i chicken out at about 70-80% of the way. Im good at managing instant losses, but i really really hate looking at profit go from +10k to +1k haha, when i get close enough to the target i either exit or place a very tight trailing stop loss. However tracking my manual trades i see in like 8/10 trades that i would have gotten more if i had held on to the stock. way past target sometimes as well. Im working on the mental part on this one hehe..
I use the “purple triangle” indicator in PRT to screen for my stocks. Ive added a “count how many triangles” indicator as well, so if it sees +5 triangles closely together it ranks it higher.
So how does that look like in real charts. please see examples. Let me know if you got any questions.
Anything im missing here?
Anything youd do different?
What does your manual system look like?
I’ll bite.
Anything youd do different?
Everything. 🙂
If I’d have to trade so “statically” as your means, I wouldn’t have a life. Meaning : When I am on the phone my trading commences. Like I write this post and may perform a couple of trades. Stupid example : see below (currently the time is 05:30). Of course you see one poor trade and a second which would be smarter, but the poor one all relates to yesterday and how that evolved (from good to super bad to perceived good again after Market close (22:00 my time) up to 23:00 to waking up this morning everything again going down the drain – meanwhile me telling to myself “Peter, you’re getting scared – don’t do that !” – with thus this as the result.
Anyway …
Trading for me is a natural habit which can be done without thinking. Observing charts ad infinitum, yes. The latter in itself is easy : always have you monitors full with charts, meanwhile doing your work like programming or talks on the phone. But anyway always.
It occurred to be that this cannot be taught or else I don’t know how to do this.
To be honest, with Trump around this has become harder (for me). It takes months and months (even after his first term) to learn or have trust in the fact that newly put due tariffs (like last week on to Europe) will be undone the next day or week. Or, how Ukraine keeps on influencing Germany (like DAX) and how that has become untrustworthy (a few years ago behavior of DAX mostly equaled behavior of US futures – this is not so any more at all).
So yes, this is very much different from preparing for today already yesterday, find the imbalances and such, and use the Volume Profile at it (which in PRT not even is possible as far as I know).
My trading habits would change over time; prior to Corona I tended to have 60 or so Stocks in my Portfolio, doing 20-40 trades a day on them. This was mostly about what to ditch and what to get back in place. But with Corona this ceased to work because all was only negative and negative stances I can’t bear. Read : I can’t go Short on Tesla really because I think they deserve no negative thoughts (though this depends of course, like Musk kissing Trump a year ago, waving his chain saw). So that’s where I started doing Futures.
So as you can see, no real strategy on this. Just trading. Having fun. Doing it for the hobby only – this is important.
if you earn money you must not ask anybody. what you do for me it looks like a strategy, that is positive, but it is extremely overcomplicated. markets are extremely complex, and if one wants to deal with them, make money there, one needs dumbass simple approach, that is my conviction, without telling details an writing 1000 explanations.
I call it my A+++ setup where i want to see at least 5/6 stars to buy:
- Price is on a support
- Price is bouncing on a trendline
- RSI broke and retest a resistance which become support
But now, as i reach 1000+ indicators, i have many mores setup 😉
jebus89 wrote: my best setup for manual trading systems
Just curious … have you tried coding and backtest your manual strategy?
If Yes, are results acceptable?
GraHal wrote: jebus89 wrote: my best setup for manual trading systems Just curious … have you tried coding and backtest your manual strategy? If Yes, are results acceptable?
I cant for the life of me code it, tried many different ways but its just not doable (edit: with my current knowledge). So what im working on currently is coding something that lets me screen for stocks that MIGHT be in the setup. Its not great results so far. i still need to go through 100+ stocks to find 1 that looks good. And that might be intuition part of it. If you look at the chart you get a hunch that “this looks right” or “this dosnt look too good”.
PeterSt wrote: ….Just trading. Having fun. Doing it for the hobby only – this is important.
I agree 🙂 (but i would also like to earn some money doing my hobby hehe)
GraHal wrote: jebus89 wrote: my best setup for manual trading systems Just curious … have you tried coding and backtest your manual strategy? If Yes, are results acceptable?
The problem would be to make it enough accurate (which demand enough data/pivots) and at the same time enough fast to execute… Looking for accurate supports/resistances, and trendlines would make the system very laggy in realtime and would need lot of time to backtest (not hours, but days for 1 backtest of 200.000 candles)
jebus89 wrote: tried many different ways but its just not doable
Have you tried getting AI to code it, might get closer than your own code?
Gemini comes with a good recommend from Ivan on here.
Yea its the trend lines, continuation patterns and a combination of it all thats really hard. I mean you basicly “got my system” there so feel free to do whatever u like with that, if u got somehting u think might work, holler at me and i can check it out and compare to what i do. You also see examples of real trades i am currently in or have taken in the screenshots
if ur AI code is taking those same trades im guessing u got close to my system
As i mentioned im currently just trying to code those purple/red/green triangles as you can see in screenshots, that should be “in the middle of a continuation pattern” in an uptrend, but they dont like “bigger trends” so i would need more indicators or larger variables etc.. its a hassle so i just mainly screen on finviz.com with very simple “the stock is going up” screen and quickly look at charts of 100 stocks per week kinda thing
I have got a question, Jebus: how do you verify if your system is long-term profitable? I mean you can’t run a backtest, and with so many complex and subjective parameters I imagine it’s close to impossible to do even objective “manual” (visual) backtest. Or was it possible for you with your particular system described?
I do not trade manually, probably since more than 10 years. I was trading manually earlier simply because I had no tools for automatic trading at that time (and my results were terrible…). still because of your topic I was thinking now under which conditions I would trade manually and only 2 good reasons came into my mind:
if I would trade purely based on fundamental ideas or
if I would have purely technical system which is not possible to code
well, trading based on fundamentals… big guys (hedgefunds etc) trade on fundamentals, so it has to be possible to make money that way, but I am a small mouse and have no solid idea how to trade based on fundamentals. and I would have my worries about long-term profitability because I have also no clue how to “backtest” such way of trading. and on top I think it is kind of dangerous to totally ignore “technicals”. so I don’t know if I ever will be trading based purely on fundamentals, but maybe combination of fundamentals+ultrasimple technicals (“technicals” like in current algos, just on higher timeframes then, daily or weekly probably).
technical system which is not possible to code… I have no ideas which I can’t code – probably because I trained my brain not create such ideas 😀
recently I discussed with a guy who trades mostly based on fundamentals but is getting excited about technical trading. he was in particular eager to trade “wolfe waves” and asked me if I could code an algo for him. I think it’s ultra-complex and subjective setup, probably not possible to code. even coding a “simple” setup like triangle is absolutely not trivial. or I maybe just lack imagination and abilities.
on other hand, if a setup is objectively not possible to code, it might be a “gold mine”.
So my journey started with manual trading and i realized 2 things:
My ideas like “this should probably go higher if breakout here” where usually correct however my emotions got the best of me and i closed trades early, i moved stop loss or targets etc. So i though well what if i could just code this and let the algotrading do it for me. Thats what got me into the algotrading.
The other thing i realized was that i love rules. I love trying to crack the markets and figgure it out. I love algotrading, but i can definitly see that some good stock picks are profitble as well. So I’ve been working on algotrading for so long, but i’ve always had this “secret passion” of wanting to crack the manual trading code as well. So i combined the two. Making rules for my manual trades to make them more systematic, but i havnt been able to actually code the trades because of the subjective “this trend looks really good”.. I need to look at the volatility of the pattern, the volume profile, the price action around support/resistance, past behaviour..
Anyways, I have tracked my manual system across 100’s of real manual trades i have done over the past couple of years, i have rated each one from 0-6 stars based on the rules mentioned above and the results was really good (!!) for the 5-6 stars trades, and it was really bad for the 1-3 stars and pretty much break even for the 4~ star trades.
As you said, a part of it is subjective and really hard to code. Like my screenshot of Berkshire stock, that trend just screams continuation. As long as its inside the trend i just wait for the continuation patterns to show up and i react on it. At the peak you can see the massive volatility spike vs the other patterns, the volume profile also screams “distribution” and no accumulation. Thats not another continuation pattern, at least not in my eyes. That dosnt mean i can correctly predict EVERY peak in a trend, but it helps me to at least miss a bunch of them..
I dont think we can compete with “the big boys”, no chance, however i do believe “everything you need to know about the market your looking at is shown in the price”.. So if big boys are going into a stock, theres ways to tell that they are, and theres ways to tell that they are not. Price action, inside buyers news, indicators etc that can help to show whats going on.
If it is couple of hundreds of trades you made according that system and you are happy with results, it’s very good base for trusting the system, does not matter how subjective the criteria are. And I can literally feel your passion in your various posts, the passion is cucial.
Regarding “what is missing”: maybe, just maybe overall market behaviour as “technical” condition, and that in some very simple way (which you can easily code), like “was nasdaq (or any other market where particular stock which you want to go long is listed) closing higher than x days ago?” I would not add this condition, but try to replace with some others, because you already have so many. all that elliott and fibonacci stuff is ultra-subjective for my taste. or: maybe some fundamentals would be of interest, again, as very simple filter: are earnings growing since x quarters? is growth of earnings accellerating? or same with turnover, in particular if company is not making money yet.
cheers
justisan
I used to try and cobble together trading systems using all sorts of indicators, hoping they’d make me money. It worked more or less… sometimes… Then someone gave me a not-so-subtle hint… I think it was you, @justisan. 😉 Since then, I see things much better and more simply, and it works. It just works. These days, I focus entirely on mean reverse in the S&P 500 or Nasdaq, and it works even with the simplest rules. You just have to filter out the extreme situations… like Covid or Trump.
😀
yep, I think it was me… some two years ago, right? it’s probably such a heavy paradox for the normal mind what super-simple approaches (no indicators, and very few variables) can acchieve that nobody is even looking in that direction, nobody thinks it can work, so nobody even tries. ok, it is not really nobody, but very few think and try in that direction. I would never ever trade the system Jebus describes, for me it is total no-go, but I don’t exclude the possibility that it might work, that more complex systems might work as well – at least for some period oft time – if applied systematically and having proper exit method. I just dont know why I would try complex stuff if stupid-simple (and easy to code) works quite perfectly. it is also my conviction that more complex systems are very sensitive to market conditions: if overall market is in strong 3-6 or more months up-trend, very different long-systems will make money, simple, complex, very complex, or even random entries with prudent exit strategy – all of them might make money (complex might make even more money than simple ones in that period). as they say in german “die Flut hebt alle Boote” (something like: the tide lifts all boats). I remember very well end of 90ies, I was moving from Lithuania for living and studying in Germany at that time, and everybody around me who had spare money was buying Neuer-Markt-stocks, and Nasdaq-stocks and signing every possible stupid IPO and cashing in like crazy, so not even having a strategy worked at that time for few years. but when market turns, conditions change, only those simple systems might survive – and still make money. for my own part, my algos don’t even have a defintion of “trend” behind the entry/trading logic. I would say majority of my systems profit from trends which last for a day or few days (rarely longer than a week) – but they don’t wait for trend showing up and enter then and then follow, instead they try to catch the trend at the moment it might emerge, disregarding completely curent overall market conditions. and so they are kind of rather insensitive to market conditions.
if I would need trade manually, I would trade same systems and same way which I now run as algos. but I think I would not be able to trade them all (9 these days) at the same time, even they trigger “only” some 3 trades per day on avarage. I think also I would die from boredom while waiting for setups to emerge.
cheers
justisan
