has anyone EVER seen a successful trader/trading system? | Page 77 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Successful Trading & Determining the Probability of a Trading System 10 replies

Has anyone ever seen this error message? 2 replies

SwissFXBroker (SFXB)--has anyone ever used them? 39 replies

Anyone ever seen a 'move to breakeven' script? 3 replies

New trading system hard to completely follow... anyone ever have this problem? 0 replies

Attachments: has anyone EVER seen a successful trader/trading system?
Exit Attachments

has anyone EVER seen a successful trader/trading system?

  • #1,521
  • Oct 26, 2020 8:12pm Oct 26, 2020 8:12pm
Quoting JamesSmithFX
{quote} For us it doesn't matter if it's manual or automated. But it's probably automated now. The question is how to detect it with our own robots? Now I want to build a robot to front run the bank robots. The bank robots are coded by people. People make mistakes in their logic. All robots have certain triggers for activation and they also may have bugs in them. Find the triggers, find the bugs. We have to Look for clues and footprints. Then we can potentially reverse engineer the logic of the bank robots.
Something like the guy who reverse engineered McDonald's data to figure out ice cream machines that weren't working and made an app, so people avoid going to those locations. I don't know how he could get access to MCD's info about ice cream machines but I think getting access to bank transactions, their robot patterns or particularly the bugs in them is going to be a lot harder. From our point of view, I think we call those transactions gaps and spikes or high volume or we just call them weird things that the pairs are doing. Our indicators are like Seismographs that are supposed to detect and act on the waves and irregularities of them. Any other way of front running them....um...suspect IMHO. lol
  • #1,522
  • Oct 26, 2020 8:16pm Oct 26, 2020 8:16pm
Quoting Wavegarrick
{quote} Here's the thing. I have been around robots for many years. You cannot expect a robot that takes counter trend trades in ranging periods to perform well in trendy periods. Vice versa, you cannot expect a trendy robot to perform well in rangy periods. The hard part is that you will never know when a range becomes a range until it is a range many candles later and vice versa when it is a trend. Can you code this into an Ea?, absolutely No. An Ea will never see what the eyes can see and even the eyes see it too late. And even if you use the...
"An Ea will never see what the eyes can see and even the eyes see it too late." I agree with you mate
  • #1,523
  • Oct 26, 2020 8:30pm Oct 26, 2020 8:30pm
Quoting OutThere
{quote} Something like the guy who reverse engineered McDonald's data to figure out ice cream machines that weren't working and made an app, so people avoid going to those locations. I don't know how he could get access to MCD's info about ice cream machines but I think getting access to bank transactions, their robot patterns or particularly the bugs in them is going to be a lot harder. From our point of view, I think we call those transactions gaps and spikes or high volume or we just call them weird things that the pairs are doing. Our indicators...
Yeah. I think we are totally screwed on this front. The only way to gain access to the necessary information is to hack the bank servers. Or to get a job there as a software developer
Anyway.
Will have to stick to the pattern recognition project and the neural networks.
  • #1,524
  • Oct 27, 2020 1:09am Oct 27, 2020 1:09am
Quoting JamesSmithFX
{quote} Yeah. I think we are totally screwed on this front. The only way to gain access to the necessary information is to hack the bank servers. Or to get a job there as a software developer Anyway. Will have to stick to the pattern recognition project and the neural networks.
Haha.
  • #1,525
  • Oct 27, 2020 5:25pm Oct 27, 2020 5:25pm
Quoting JamesSmithFX
{quote} Yeah. I think we are totally screwed on this front. The only way to gain access to the necessary information is to hack the bank servers. Or to get a job there as a software developer Anyway. Will have to stick to the pattern recognition project and the neural networks.

All the information we need is right on the charts. What the banks are doing leaves a trail on the charts.
If we look well and long enough, we'll be able to connect the dots. OHLC, Correlations, Co-integration...
You will figure out a lot of statistics with AI if you feed it with relevant data. Lots of feature engineering. An understanding of the fact that the price of each Currency pair is dependent on what others are doing...

But then, there's the LP trading against you. Yes. If you're making money, they will soon start trading against you.
That's what makes RickM's trading style of interest.
If you can code it...
  • #1,526
  • Edited 6:38pm Oct 27, 2020 6:27pm | Edited 6:38pm
Quoting peterbabs
{quote} All the information we need is right on the charts. What the banks are doing leaves a trail on the charts. If we look well and long enough, we'll be able to connect the dots. OHLC, Correlations, Co-integration... You will figure out a lot of statistics with AI if you feed it with relevant data. Lots of feature engineering. An understanding of the fact that the price of each Currency pair is dependent on what others are doing... But then, there's the LP trading against you. Yes. If you're making money, they will soon start trading against...
I don't know about this. I think that the retail traders imagine things. I see lots of delusions here. Lots of wishful thinking.

I look at the history and so far the charts look completely random. I can't see any real tradable patterns. The whole "price action" looks like some noise created by random process. Like Brownian motion.
Actually, there is only one visible and recurring pattern and this is the contraction and expansion of the intraday volatility. It's very easy to see it with tick volume or atr indicator on H1 chart. Other than that I can't see anything else.

So if there are some other patterns they are hidden very well in the data. If they are hidden so well this means that the patterns are very weak and it's almost impossible to make money from them. The expectancy per trade will be tiny. Probably not enough to cover the spread.

Now the only thing left is to build my AI Neural network and to proof all of this.
  • #1,527
  • Oct 27, 2020 6:58pm Oct 27, 2020 6:58pm
Quoting JamesSmithFX
{quote} Now the only thing left is to build my AI Neural network and to proof all of this.
that moments of catching the easy 10 - 20 pips are on the charts. I am ashamed to tell you that it requires only a few lines of code for the EA and does not need a computer scientist to code million of lines.
  • #1,528
  • Oct 27, 2020 7:10pm Oct 27, 2020 7:10pm
Simply said, if price does not want to go in one direction then it will go in the opposite direction. As a trader you must look for the signs that price is showing you which direction it wants to go.
If you don't want to look for level rejections then look for trend rejection or pop the pimple and look for both.
Place a medium range moving average on your chart and you will see price follow that moving average- then price will change its direction and go towards the MA. If price then returns back in the original direction as the MA then it is telling you to jump on board. Note how many time price will move in to visit a moving average area and then move back away from the moving average, this is a signal for entry especially if price leaves a PA pattern or rejection from the MA area.
If you don't want to use a moving average then use support or resistance areas as they do the same thing.
Sometimes I think the biggest mistake for a trader is to analysis to many charts or to many pairs- stay with the higher time frames and allow yourself enough to analysis what the future setups will show.
Don't chase quantity, look for quality.
  • #1,529
  • Edited 6:38am Oct 28, 2020 6:24am | Edited 6:38am
Quoting peterhenders
Simply said, if price does not want to go in one direction then it will go in the opposite direction. As a trader you must look for the signs that price is showing you which direction it wants to go. If you don't want to look for level rejections then look for trend rejection or pop the pimple and look for both. Place a medium range moving average on your chart and you will see price follow that moving average- then price will change its direction and go towards the MA. If price then returns back in the original direction as the MA then it is telling...
Hi Peter

I see the market as just a bunch of strong passive limit orders that get attacked by weak aggressive market orders -

NO trends, NO resistance areas, NO support, NO demand or any other rubbish concept.

Price for me goes in the direction of the largest density pool of orders. We are talking about passive limit orders which tends to cluster together in pools of liquidity because traders think alike, they move in herds.
So price is fluid and dynamic and just moves towards these of weight of orders always as the passive orders are entered on to the market.
We need aggressive Buy market orders to start the push up then if we want this move to continue, buyers need to disappear.

It sounds strange that we need buyers to thin out on a Bullish market but for this move to continue, we need sellers to enter with their stops. What I see as I watch this action unfold in front of my eyes on the M1 chart with the tools in front of me is the big players are feeding buy orders directly to these sellers but in a matching density.

Finally aggressive Sell orders overwhelm the big players BUY density so price collapses. Therefore it is not volume that's important, rather how dense the liquity pool has become.

So the best chance of profitability is fade the buyers or sellers as they leave the market just before the aggressive market players enter.

Cheers
Attached Image (click to enlarge)
Click to Enlarge Name: order flow graph.png Size: 264 KB
Trading thin liquidity at the boundary of the charts
  • #1,530
  • Oct 28, 2020 7:14am Oct 28, 2020 7:14am
Yes, supply and demand is what makes the price move. But you don't need "aggressive market orders" for price change to occur.

I think the price changes with the limit orders at the top of the book.

What happens if there is some news and all limit orders are repriced instantly 100 pips higher? I think the new current price will be 100 pips higher without any market orders executed inside the gap. Is this correct?

I think most of the larger movement happens like this. It's not because of marker orders. It's the market makers (aka liquidity providers) repricing their limit orders up and down all the time. It's artificial movement created by algorithms.

Fundamentally It works just like any other business. For example, If you have a shop and you price the goods too high they won't sell. You have to reprice them lower. But if you price them too low, you will lose money. This is the simple supply and demand. You have to search for the balance.
  • #1,531
  • Oct 28, 2020 8:03am Oct 28, 2020 8:03am
Quoting JamesSmithFX
Yes, supply and demand is what makes the price move. But you don't need "aggressive market orders" for price change to occur. I think the price changes with the limit orders at the top of the book. What happens if there is some news and all limit orders are repriced instantly 100 pips higher? I think the new current price will be 100 pips higher without any market orders executed inside the gap. Is this correct? I think most of the larger movement happens like this. It's not because of marker orders. It's the market makers (aka liquidity providers)...
It never happens, I think you need to study the charts for a while longer.
A move may run 70-80 pips in a few seconds like it did last week - but every single move collapses. The liquidity becomes barren, void of any useable orders so price withdraws to density.
I mean no offence but you need to study how liquidity moves before making those statements James.
Trading thin liquidity at the boundary of the charts
  • #1,532
  • Oct 28, 2020 8:20am Oct 28, 2020 8:20am
Quoting RickM
{quote} It never happens, I think you need to study the charts for a while longer. A move may run 70-80 pips in a few seconds like it did last week - but every single move collapses. The liquidity becomes barren, void of any useable orders so price withdraws to density. I mean no offence but you need to study how liquidity moves before making those statements James.
I know how the order books work. I buy stocks every month. Real stocks traded on exchanges like NYSE, NASDAQ, LSE and many other.
I know that when your add or remove limit orders at the top in the order book the price changes.
  • #1,533
  • Oct 28, 2020 10:14am Oct 28, 2020 10:14am
An EA can be successfull, BUT you need to treat it like someone who is doing the dirty work for you and guide and improve it all the time.. I don't believe in those 'plug and play' ea's that are running the same way all day 24/7 and certainly not those that use hedging, grid, martingale etc etc.. creating your own ea is the fastest way to becoming profitable, but remember you need to guide it constantly.
  • #1,534
  • Edited 11:11am Oct 28, 2020 10:39am | Edited 11:11am
Quoting mtruefx
An EA can be successfull, BUT you need to treat it like someone who is doing the dirty work for you and guide and improve it all the time.. I don't believe in those 'plug and play' ea's that are running the same way all day 24/7 and certainly not those that use hedging, grid, martingale etc etc.. creating your own ea is the fastest way to becoming profitable, but remember you need to guide it constantly.
If you have to "guide" the EA constantly this means that the EA is primitive and incomplete.
To evolve further, we should focus on the "guiding process" and find ways to automate this as well.

If we make a breakdown of the guiding process what do we see?

1. Analysis of the results and the market over the last N bars.
2. Adjustments of the parameters or/and adding of new rules.

Both of these activities can be automated with integrated machine learning.

In fact, if your integrated machine learning system is very good then your can start with blank EA without any trading rules.
Or basic rules like: if(LongPosition==0 && Price[0]>Price[1])OpenLongPosition();

After some time the EA will start to learn from the trade history to create it's own rules. The EA will learn from the mistakes and will start to modify the rules until maximum capacity is reached.
  • #1,535
  • Oct 28, 2020 11:38am Oct 28, 2020 11:38am
Quoting JamesSmithFX
{quote} I don't know about this. I think that the retail traders imagine things. I see lots of delusions here. Lots of wishful thinking. I look at the history and so far the charts look completely random. I can't see any real tradable patterns. The whole "price action" looks like some noise created by random process. Like Brownian motion. Actually, there is only one visible and recurring pattern and this is the contraction and expansion of the intraday volatility. It's very easy to see it with tick volume or atr indicator on H1 chart. Other than...
maybe you shouldn’t even look for patterns, but something completely different.
  • #1,536
  • Oct 28, 2020 11:47am Oct 28, 2020 11:47am
Quoting tiborf71
{quote} maybe you shouldn’t even look for patterns, but something completely different.
And what is that?

News?
Macroeconomic trends?
Central bank policy and interest rates?

I thought about these but....None of this works.
The big fundamentals are priced in advance and the news are simply untradeable because the market gaps and moves too fast.
  • #1,537
  • Oct 28, 2020 12:01pm Oct 28, 2020 12:01pm
Quoting JamesSmithFX
{quote} And what is that? News? Macroeconomic trends? Central bank policy and interest rates? I thought about these but....None of this works. The big fundamentals are priced in advance and the news are simply untradeable because the market gaps and moves too fast.
try trend trading, if you study this opportunity, you will see that there is potential in it if you act smart.
if you do it right, you can beat the market.
it matter here whether the market is random or not because trends exist.
  • #1,538
  • Oct 28, 2020 12:20pm Oct 28, 2020 12:20pm
Quoting JamesSmithFX
{quote} I don't know about this. I think that the retail traders imagine things. I see lots of delusions here. Lots of wishful thinking. I look at the history and so far the charts look completely random. I can't see any real tradable patterns. The whole "price action" looks like some noise created by random process. Like Brownian motion. Actually, there is only one visible and recurring pattern and this is the contraction and expansion of the intraday volatility. It's very easy to see it with tick volume or atr indicator on H1 chart. Other than...

I can tell that you haven't spent enough time looking at the charts.

Wish you success with your AI.
  • #1,539
  • Oct 28, 2020 12:23pm Oct 28, 2020 12:23pm
The most successful traders are the people you're trading against. Think about it. They charge a spread, a transaction fee, they know where your stops and targets are AND they don't give us that data. What a great system. It's amazing they have any losing days, but they have to keep the fish swimming so they have to feed them now and then. That is my answer for the threads original question.

So you can't charge a spread or a fee to them, but you can look at your chart and figure out where all the other traders stops are. Why, because the banksters are greedy beyond understanding and that is where the easy money is. They like easy. Spread plus fees plus profit equals big money for them spread out over many fish.

Some people may not want to hear what I'm saying but after 15 years of watching charts I've seen too many things to unsee them. It's all about liquidity and where the easiest money for them is located. That's how their algo's are designed. There are No rules. Think herding cats. But they want you to think in those terms like Elliot or highs and lows, etc....and the one thing they always insist everyone should do is use a stop. Hmmmmm. That's handy. For them.

Time is the key. When!!! No one discusses time. All but a couple of tools are price based, almost none time based. Why is that? There are two parameters on your chart. Price and time. Price is all everyone is concerned with. Jesse Livermore once was quoted something to the effect " Tell me when and I'll be rich". Figure out when and then you have something.
  • #1,540
  • Oct 28, 2020 1:40pm Oct 28, 2020 1:40pm
Quoting tiborf71
{quote} try trend trading, if you study this opportunity, you will see that there is potential in it if you act smart. if you do it right, you can beat the market. it matter here whether the market is random or not because trends exist.
so are you using some TF strategies that still work in 2020 ?
it seems to me that TF doesn't work that well on FX/metals anymore.
maybe I'm wrong... do you have a myfxbook or fxblue account ?
0 traders viewing now