ReadtheMarket - Supply and Demand with Price Action

Hello ahdat,

I wanted to send you a pm, but since you are a new member, I can't. I am very interested in how you managed to accomplish the automation you are speaking of.

I am trying to do the same, just started studying ACSIL in Sierra Chart, as that is my platform. The truth is I can manually find many exceptional no-drawdown entries, grateful to IF for all this free knowledge, however I invariably mismanage the trade and exit wayyyyy too early (for example, having a target of 20R and exiting at 2R. Most of the time the original target is eventually hit.)

This is based on fear of loss/impatience, which of course produces the losses we are fearful of (and leads to overtrading and blown accounts). I can see no other solution than technology helping me, but I can't see how this method can be completely automated.

I have ideas on how to define engulfs and FTRs, and use multiple TF charts, to help combine and compare the different scales. I have also developed some ideas, mostly based on liquidity gaps. Also, I'm hoping to use previous swing extremes to use their stops as targets.

However, I can't see how this could be completely automated. I am hoping it can greatly reduce the decisions I have to make, and so reduce the stress, that is just too much.

Complete automation would certainly be much less profitable than a skilled and calm manual trader (e.g. IF), but-for me at least- would be a godsend at this point in time. If I am at the monitor, I will mess the trade up, no doubt about it.

I would greatly appreciate a few pointers from you, either here or at alphagad79 at gmail.com, and maybe we can exchange ideas.

Thanks a lot and good trading!

hey alphagad,

I went the automated route primarily because Pacific time has to be the absolute worse place to be for trading if you want any semblance of a normal schedule. I think its been the number one thing holding me back as i'm fighting sleep and life schedules on top of learning to trade. Plus i'm naturally a much better coder than 'trader'; I deeply enjoy the coding and analysis. It was the right path for me.

How much experience do you have coding and how much do you enjoy coding? My platform started as just a 'helper' to identify trades, and morphed towards full automation as things naturally started coming together. As you can imagine, its pretty damn complex, implementing something at this level would require a fairly high degree of coding experience and RTM knowledge (or a lot of time to learn). In other words, don't take this journey lightly. Expect this to take 'months' to get to a barely functional state.

Anyways, some hints:

1. Take it one step at a time. Start by building a 'helper'. Something that visualizes the charts for you (you'll need this anyways to evaluate your robot's analysis). You'll get a feel for how good your ideas are and how good your robot is at reading the charts and identifying trades. Only start thinking about full automation after you get this part down.

2. Right now i use the 'zigzag' indicator as the foundation of my robot. I might make a custom algorithm at some point, but its sufficed for now. This adds another layer of complexity integrating different zigzag settings along with different timeframes.

3. I'm still working on it to get more consistency and reduce its risk so it won't blow my account. This is probably the most important factor. There are days and scenarios when my robot does extremely well, but it only has to blow my account once for it to be game over.

Hi Nehru
As for price's reaction on 1.32600-1.32800, I would respectfully disagree with that. Unless (lets calls it) YourTF (the timeframe you derived the zone) has the most smallest zone in that area and therefore, most accurate zone. I am not sure if that is where the pending orders are. I've read, if I remember correctly, Fxyogi/ IF say that the timeframes are just an illusion. HTF shows the strength of the move, but LTF is where we can pinpoint exactly where the orders are. And that is what I tried to do here, but it didn't react as I thought it would. If only IF and the seniors can enlighten me if I lacked something in my analysis in this portion of the chart. It is either I lacked something in my analysis or my FL wrong.

As for price's reaction around 1.33000, I have doubts on the FL you drew. I am not the authority on Fls, but this is my view.


Regardless, I think you have a point here. Thats what Chica was probably referring to. Talking about Chica... I found something funny.

Thanks ahdat for the insight. I have a lot of RTM and price knowledge, I'm just useless at coding, just starting. I understand this will be very complex, at least if it's to be effective. That's why I'm not envisioning, at this stage, full automation but a helper, as you put it.

I tried to make do with the spreadsheet function of Sierra Chart, but it just won't cut it for something this complicated. I'm sure I will enjoy coding, it will definitely take a lot of time though. It's based on C++ too, hopefully I'll make it. Do you use MT4 for your trading?

A couple more questions, if you don't mind, and I'll try not to bug you again.
1) Do you find that zigzag works better than a swing high/low indicator? One uses a fixed number of ticks, the other uses time/number of bars. A combination of these would be ideal, but I haven't seen anything like it. I'm thinking of using the swing indicator and logic to see if price crossed it (with different settings for HTF swings), as the zigzag doesn't care about breaks of highs and lows.
2) Do you use LTF PA confirmation (e.g. M1 engulf of the base that made the high; base, that will be a nightmare to program lol) for your entries, or just input entry price/targets after an area has been identified by the automation logic?
3) Do you employ a risk management logic to filter simultaneous setups and pick the "best" or stick to a couple of pairs?

Thanks again!

Hee hee!

I've just looked at a thread called "Noise free systems"

Sounded good. No noise, just price, right?



Wrong!!!





The trading World needs to be saved from itself!

Doing so, you're attracting a new horde of haters and trolls!
Prepare for attack!

(Nice colors btw)

Till RTM is active again i will post here ....

Another Trade still need to learn more ...

It's more like this...

I've said that I wouldn't post here anymore. Only fools do not change their minds

alphagad,

You got a good attitude, glad to answer any questions. I use FXCM's platform, but i'll probably port it to something else at some point to give myself more options.

1. I'm not familiar with the swing indicator but i would imagine anything that identifies swings would work. The zigzag actually does a pretty good job identifying the highs and lows. But you pointed out the limitation of zigzag. Different settings will give you very different zigzag results. But sometimes thats actually a good thing because you can finetune specific things you're looking for. For example, a very low depth will help you identify significant lq gaps. maybe i'll take a look at the swing indicator today

2. Right now no ltf confirmation, but that's on the list. My idea was to take into account several factors to determine how to enter a trade. Analyze the approach, determine the zone's 'score', and make a determination on how to enter. Whether that's a touchtrade, engulf of last low (or last fl), etc... I think of it as building blocks. Every RTM thing (fl, key level, compression, lqgap, etc...) is a block (or in programming terms, an object in object-oriented programming). Then you can use these diffferent blocks in different ways to build and score your patterns for various uses.

A simple engulf should be relatively easy to do with any swing indicator. You have the reaction move after touching your zone. The next swing, price can't break your zone. The swing after that has to engulf your first reaction move. Then just wait for a retrace to enter.

3. I'm just running a handful of pairs. Different pairs have different personalities, and i adjust my robot for each pair. Some things work better than others, and some pairs i dont trade at all. I do filter out very similar trades for the same pair. If you're talking about managing risk across the same currency (ie, usd/jpy, aud/usd, usd/cad, etc....), that hasn't really been a problem. I'd rather increase my lot size if i see confluence.

Thank you very much ahdat, you have given me a lot to think about! I'll probably re-read this quite a few times

As for platforms, I wholeheartedly suggest Sierra, I think it's by far the best out there, and also works with FXCM, if I'm not mistaken. Thanks again.

well...

You're welcome! Good luck and have fun with it. I may check out Sierra.

For some inspiration, here's a snapshot of just the Flag Limit indicator in action in the h1 timeframe. By analyzing manually, you can see why some worked, some didn't work (and could have been avoided). So we go back and add more rules and filters and ways to analyze price through different timeframes to get things more concise. Rinse and repeat.

any expert, can you tell me why the price didnt go down after engulf FL?
thanks

Right where price bounced is an s/r flip level (extend that box you drew near the bottom left), and a FL. And price after the engulf has pretty much all been taken out, so price is likely to cut up and through.

thank you ahdat and mertch, I was too focused with this