Is random really random? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Is it just me or is cable acting random? 49 replies

MTS Vs Random Entry - A Paradox 6 replies

The Really Really Boring Forex System 97 replies

Price action pattern, and random thoughts 6 replies

Random number generation 7 replies

Attachments: Is random really random?
Exit Attachments

Is random really random?

  • #21
  • Jul 24, 2014 1:23am Jul 24, 2014 1:23am
Quoting Proximus
{quote} Depends also on the importance of the event.Lemme tell you 2 events and you tell me the difference between them: 1) Due to a landslide/earthquake a major australian gold mine just collapsed...investors are concerned over the effects of this on the gold markets 2) Australian new home sales went up 0.1% compared to previous month data Which will have a greater impact on the spot-gold markets? ..... If you answered 1) then you are correct. Besides you dont have to predict these events, just be in touch with the latest news, and trade the aftermath...
I guess you got me there, never really thought about a hypothetical situation like that. What happens if they took investments from gold and invest it in "Australia's" Booming housing market Lol. Just kidding. I get the picture, but the events I am talking about are specific in the market and happen about 3 times a week naturally in the forex market. And those are not even the only ones just good situations that put you in the market running with the wind at your back.
Being a trader is lonely, but being a great trader is lonelier still
  • #22
  • Jul 24, 2014 1:30am Jul 24, 2014 1:30am
Quoting victor_ray
I do find it's intersting some how in this world... "There seem too many people who know the news before it even publish out in online..." I alway feel like I am alway the last to know (aka price priced in). Even when it's clearly "economic calendar" stated 5 min more until news data release so either there is some B.S insider trading or I am just getting the short end of the stick everytime when it come to acquiring news before live release. However the "aftermath" is also problem. Everyone know that during 2008 Market housing bubble crash.. It...

Lol You are a funny guy Victor. May the pip be with you.
Being a trader is lonely, but being a great trader is lonelier still
  • #23
  • Jul 24, 2014 1:31am Jul 24, 2014 1:31am
Quoting Proximus
Even so if you find a temporary pattern which shows a real edge with positive expectancy trading it, you never know when those rules change.
If the rules change, the Pattern no longer exists therefore, you'd never see it again. A real Pattern has a cause behind it. If you see it, the rules are intact by definition.

Quoting Proximus
Suppose the "morning star" pattern worked perfectly 50 years ago on the S&P 500, it predicted every single reversal for 5 years with 100% accuracy.
The "Morning Star" is not a Market Pattern. It is a Candle Pattern. Market Patterns come from derivative market data, not the market directly. This is why you have to build "tools" to distinguish the potential for a Pattern and then correlate that data with actual market movement at the time the trigger is set within the tool you developed. Remember, real Market Patterns don't fall out of the sky and appear on your charts all nice and neat. If it were that easy, everybody would be doing it.

You have to hunt for market Patterns through tools that you build on your own - or become intimately familiar with tools (indicators) that someone else has built and made public. Then you correlate triggers in the tool with actual market movements to validate whether or not a quality Pattern exists. They don't just jump out and say "here - I'm a pattern trade me for profit." You have to research them, painstakingly often times.


Quoting Proximus
But then something changed, the market became more volatile and whipsawy so after 5 years it's magic dissapeared like if it never worked.
Multidimensional Time Frame Analysis and Historical Validation must be included in EVERY pattern traded - or I don't trade it. That reduces my risk tremendously and prevents the vast majority of hard losses. Whipsaws come in every market - get used to it. That's why you trade within a real Pattern. You allow the Pattern to bring price back to where it belongs. If the Pattern was researched and developed correctly, this will happen far more than it does not. I've never encountered a 100% efficient system.

Quoting Proximus
You see a random market can also produce a pattern which repeats itself, but its not actually a pattern its only a random structure which just happen to have a 5-6 sigma deviation event of repeating itself unusually many times with perfect accuracy.
A "Pattern" that is not a "Pattern," is logic that argues against itself by definition.

A Pattern exists when it repeats with sufficient historicity that yields a mathematical probability for the next occurrence within a highly probable sequence and where said probability is sufficient to meet the profit requirements of the trader. Patterns do not have to remain profitable indefinitely, in order to qualify as trade worth. Their Shape need only demonstrate a historicity for remaining intact long enough to extract a predefined profit level according to the Trader's personal requirements. Those types of Patterns are quite different from Candlesticks as they involved Structure that spans Multiple Time Frames and passes the historicity test.

Without the proper tools to search for search for such Patterns, they will be very hard if not impossible to even see, let alone evaluate.

Quoting Proximus
The problem is that it was never a pattern, it only in your mind, it's like if a european tries to read chinese text, he sees only random pixels, while the chinese guy can extract information from those seemingly random pixels which make up that character. But the fact remains that every picture/character/image/ or even price chart is just random if you look at it from your eye. However if you dont look at the candles themselves but at the context, then it may have a meaning after all, but the fact remains that those doji or morning stars remain...
When prices moves through the structure of a Pattern to form a Shape on a two dimensional chart that was previously predicted, that's a Pattern. When that same scenario happens greater than 90% of the time for that particular Shape, that's a Pattern. When that Shape lasts in the market for X amount of Time, that's a Pattern. And, when you can consistently enter a position and then wait for the Market to produce a specific two-dimensional Shape of Price on a chart, that's a Pattern. There are many Patterns in existence all the time. Those that are actually trade worthy depends on the needs of the Trader and the stability of the Pattern, which itself is determined by the inputs that give rise to the Pattern's existence.

If the currency markets were truly random, I would not be a full-time trader on an institutional trading platform - I'd be doing something different for a living as it would be pointless to trade based on my profit requirements which are fairly high. Unless you wanted to sit around and wait for News and Economic Reports and then guess about a continuation of the initial impulse from the event, I would suggest start developing tools that no one else has, look for patterns within those tools and then correlate those patterns to actual market movement for validation. Then do a lot of historical testing to really begin to understand the limits of the pattern and the optimal profit and risk levels.
F-X-C-M | A Classic Retail Bucket Shop Scam Artist Who Manipulates Prices
  • #24
  • Jul 24, 2014 1:40am Jul 24, 2014 1:40am
Quoting 60minuteman
{quote} lots of theory and no proof... the research on TA has already been done.
I don't recall being polled about my private research. "The research." What's that look like? Did it include my data models or my tools?


Quoting 60minuteman
And another thing, I dont need to develop any TA tools as trading intraday requires no TA.
Of course, not. Who said it did? I used to guess all day long, too - when I was green and completely unaware of the importance of mathematical modeling. I guess that if I were to feel a Vegas Moment coming on, I could drop a quarter or two during an NFP report and then "let it ride" just to see if the roulette wheel lands on a good number. But, then again, I would not have the expectation that I am actually trading with any degree of historical relevance.

I would be winging it and it would feel good, because I would be having a Vegas Moment. There is nothing wrong with having a Vegas Moment, just as long as I don't make that a permanent part of my business as a private fund manager of my own capital.

In fact, I'm feeling a Vegas Baby Moment coming on right now. I think I'll drop a quarter to play the JPY and the Japanese Housing Reports! Oh, wait! Those are not due quite yet. I'll have to wait - but maybe my Vegas Baby Moment will be over by then. We'll just have to see! If not, maybe I could still catch a magic show or something before getting back to Mathematical Model Driven Trading.
F-X-C-M | A Classic Retail Bucket Shop Scam Artist Who Manipulates Prices
  • #25
  • Edited 3:38am Jul 24, 2014 1:51am | Edited 3:38am
Is random really random ?

Thank you for asking the right question!

Here is the answer:

==> Randomness is proportional to the degree of observation.

and of course it does not mean that finding some order and pattern after careful observation makes the data less random!
Tthat is why Technical Analysis in retail forex trading is gambling and addictive.
  • #26
  • Jul 24, 2014 1:55am Jul 24, 2014 1:55am
Quoting Proximus
TA alone is not enough.
I'm not disagreeing with you. I just don't see how it is possible to prove such a statement.

Quoting 60minuteman
lots of theory and no proof... the research on TA has already been done.
Can you point us to some books or research papers (and preferably some that have never been disputed)?
No offense, but otherwise it seems that you've provided even less proof than the member you've cited.

Quoting 60minuteman
I dont need to develop any TA tools as trading intraday requires no TA
A couple of questions, if I may:
What is your definition of TA?
What other tools should a retail trader use?
  • #27
  • Jul 24, 2014 2:03am Jul 24, 2014 2:03am
Quoting hanover
{quote} I'm not disagreeing with you. I just don't see how it is possible to prove such a statement.
TA alone is not enough.Drawing a few nice lines on the chart wont move a 5 trillion dollar market neither it will predict it.

TA only gives a framework in which you can move, so to say you detect a major S and a major R line in which the price is confined, but that wont predict which one the price will hit first.

As a very wise trader once said: "Technical Analysis is like driving a car by looking at the rear-view mirror"

Just think about it, do you drive your car like that?

Looking at past data can only tell where the price has been, but it wont tell where it will go
"There's a sucker born every minute" - P.T. Barnum
  • #28
  • Jul 24, 2014 2:05am Jul 24, 2014 2:05am
Quoting hanover
{quote} I'm not disagreeing with you. I just don't see how it is possible to prove such a statement. {quote} Can you point us to some books or research papers (and preferably some that have never been disputed)? No offense, but otherwise it seems that you've provided even less proof than the member you've cited. {quote} A couple of questions, if I may: What is your definition of TA? What other tools should a retail trader use?
try google and your own results....
tradewith60
  • #29
  • Jul 24, 2014 2:11am Jul 24, 2014 2:11am
Quoting hanover
{quote} I'm not disagreeing with you. I just don't see how it is possible to prove such a statement. {quote} Can you point us to some books or research papers (and preferably some that have never been disputed)? No offense, but otherwise it seems that you've provided even less proof than the member you've cited. {quote} A couple of questions, if I may: What is your definition of TA? What other tools should a retail trader use?
theres so much of this out there...

http://www.stockopedia.com/content/t...harting-63806/
tradewith60
  • #30
  • Jul 24, 2014 2:54am Jul 24, 2014 2:54am
@60minuteman,
@Proximus:
I did some quick research of my own like you (60mm) suggested, seems that opinion is divided as to the usefulness of TA. Don't want to get into a prolonged debate over something which I don't believe can ever really be proven or disproven, hence I'll exit here.

But if I ever see either of you post a price chart again, I'll assume that you've changed your views.
  • #31
  • Jul 24, 2014 3:15am Jul 24, 2014 3:15am
Quoting hanover
@60minuteman, @Proximus: I did some quick research of my own like you (60mm) suggested, seems that opinion is divided as to the usefulness of TA. Don't want to get into a prolonged debate over something which I don't believe can ever really be proven or disproven, hence I'll exit here. But if I ever see either of you post a price chart again, I'll assume that you've changed your views.
doh.... thats the death of my h4power thread then

always nice butting headz with you sir
tradewith60
  • #32
  • Jul 24, 2014 4:01am Jul 24, 2014 4:01am
Quoting 60minuteman
{quote} doh.... thats the death of my h4power thread then always nice butting headz with you sir
I guess it depends on how we define TA. For the record, wikipedia defines it as a "security analysis methodology for forecasting the direction of prices through the study of past market data, primarily price and volume".

The funny thing (IMHO) is that both you and Prox are saying the same thing ("TA doesn't work"), but — at least based on the way you use charts to make prognostications in your H4 Power thread — his definition of TA is apparently more onerous than yours.

That's a problem with so many of these threads, people's differing ideas of terms like "TA" (e.g. where does one draw the boundary between TA and PA, or is PA a subset of TA?); "random" (e.g. does it mean a "totally random walk" or merely something that can't be predicted with 100% certainty); "MM" (does it include exits, or is it limited to position sizing and account/portfolio risk?); and so on, and so on.

The result is that sometimes people probably end up arguing about things that they actually agree upon, or vice versa. LOL

It probably seems that I believe TA works. The truth is I have no idea, but I doubt that it's provable, because there are an infinite number of ways that charts can potentially be interpreted, and an infinite number of tests that could potentially be run.

At the end of the day, I'm a retired programmer, not a pro trader, so my opinion is worth nothing. But when I'm in the mood, I enjoy forum debates, and sometimes playing devil's advocate just for the sake of it.

Much respect to you also. May the forex gods be with you.
  • #33
  • Jul 24, 2014 4:04am Jul 24, 2014 4:04am
Quoting Proximus
{quote} Even so if you find a temporary pattern which shows a real edge with positive expectancy trading it, you never know when those rules change. Suppose the "morning star" pattern worked perfectly 50 years ago on the S&P 500, it predicted every single reversal for 5 years with 100% accuracy.But then something changed, the market became more volatile and whipsawy so after 5 years it's magic dissapeared like if it never worked. But if you still trade it after 5 years, until you find out that its not working anymore you could potentially lose all...
Yes, but lets look at it like this...the pattern itself is formed from a price, so lets say that one day you have a huge sell-off of cotton, and the price is pushed to extreme lows for that day...then you have another day where no one wants to sell at that price because they think that the price is too low (you don't want to sell your hard picked cotton by your hard workers at that price because you have to pay them salaries, and you have to make for a living), so the price stalls for that day...the next day there are people who bought cotton at that low price two days ago and now they want to sell it higher and at the same time there are people that want to buy cotton at that higher price and there you have it, price suddenly starts to gain new highs, and forms your "morning star pattern", not because they all agreed to do that but because they saw the opportunity to buy something cheap...in my studying of candlestick patterns and TA in general I saw that the patterns themselves are not the cause of something but the effect... A lot of people treat them like a cause....
  • #34
  • Jul 24, 2014 4:10am Jul 24, 2014 4:10am
Quoting hanover
@60minuteman, @Proximus: I did some quick research of my own like you (60mm) suggested, seems that opinion is divided as to the usefulness of TA. Don't want to get into a prolonged debate over something which I don't believe can ever really be proven or disproven, hence I'll exit here. But if I ever see either of you post a price chart again, I'll assume that you've changed your views.
Man you clearly didnt read my post.I didnt said that TA is useless, it has its own use, which is to determine the frame in which to trade, and to navigate trough the price in an orderly way.

The only thing that i disagree about TA is its predicting capability, like you see guys using MACD & Stochastic and they naively think that that will predict the price.In that case i totally reject it.

However basic S&R, ocasionally trendlines, but i dont really like them, and thats all.I use 1 single indicator at the moment which i myself developed, and thats about it.But i dont trade off technicals solely anymore, that was my biggest mistake that i did in trading.
"There's a sucker born every minute" - P.T. Barnum
  • #35
  • Jul 24, 2014 4:12am Jul 24, 2014 4:12am
Quoting merquise
{quote} Yes, but lets look at it like this...the pattern itself is formed from a price, so lets say that one day you have a huge sell-off of cotton, and the price is pushed to extreme lows for that day...then you have another day where no one wants to sell at that price because they think that the price is too low (you don't want to sell your hard picked cotton by your hard workers at that price because you have to pay them salaries, and you have to make for a living), so the price stalls for that day...the next day there are people who bought cotton...
Yes it is the effect, but you cannot predict the price from the effect, you can only predict it from the cause, find the cause and then trade it accordingly and dont waste time analyse patterns after.
"There's a sucker born every minute" - P.T. Barnum
  • #36
  • Jul 24, 2014 4:16am Jul 24, 2014 4:16am
I am practicing to trade incomplete patterns...I complete them in my head..
  • #37
  • Jul 24, 2014 4:23am Jul 24, 2014 4:23am
Quoting merquise
I am practicing to trade incomplete patterns...I complete them in my head..
How do you know that that "half pattern" of yours is caused by the same thing as the previous one which occured last time?

Since if not then there is a good chance of different market reaction, maybe a whipsaw, maybe a total opposite reaction.

It's like the NZD/JPY massive selloff after the RBNZ increased the interest rates by 0.25%.Everyone was expecting extreme bullish reactions since a rate hike is bullish no ? Well those newbies have totally failed to look at other factors too, i was bearish on in for 2 days, and after i saw the rate hike i entered immediately short, and now its about 46 pips down in profit.

You see the same interest rake hike which should be bullish, had totally opposite reactions.

How do you foresee that with your incomplete pattern theory ?
"There's a sucker born every minute" - P.T. Barnum
  • #38
  • Jul 24, 2014 4:25am Jul 24, 2014 4:25am
Quoting iDoubleStoch
{quote} Study the Weather/Meteorolgy. You will find exactly what you just alluded to within the earth's atmosphere happening all the time someone on the planet's surface. Weak Cold Fronts or Occluded Fronts where a cold front overtakes a warm front produces many cloud patterns that are genuine clouds, but that do not have much stability. Those clouds will be rapidly changing their Shape. What you want are market Patterns that are the most resistant to changes in their Shape over Time. Shape over Time. That's a big key. You are looking for the Shapes...

Thanks for the advice, but I already have my edge. What i am trying to do now is turn it into code, the good thing about it though is it doesn't use patterns like most other methods that i see. My method follows the flow of money throughout the entire market where ever it leads and pairs that move more freely than others.
Being a trader is lonely, but being a great trader is lonelier still
  • #39
  • Jul 24, 2014 4:30am Jul 24, 2014 4:30am
Quoting Proximus
{quote}But i dont trade off technicals solely anymore, that was my biggest mistake that i did in trading.
Good for you man, that's what I did a while back. And it's the single biggest reason why so many are only talking about S/R and that kinda stuff because they don't want to admit that there is something missing and they go on doing the same thing. They try to trade Forex like they do for stocks or the likes, but they don't realize is that a currency gets it value 3 ways, and that gives an edge to match it up when any currency that is best at any given time.

It can be done with TA, but that's like trying ti cut grass with a pair of scissors. But to each his own I guess.
Being a trader is lonely, but being a great trader is lonelier still
  • #40
  • Jul 24, 2014 4:30am Jul 24, 2014 4:30am
Quoting Proximus
{quote} How do you know that that "half pattern" of yours is caused by the same thing as the previous one which occured last time? Since if not then there is a good chance of different market reaction, maybe a whipsaw, maybe a total opposite reaction. It's like the NZD/JPY massive selloff after the RBNZ increased the interest rates by 0.25%.Everyone was expecting extreme bullish reactions since a rate hike is bullish no ? Well those newbies have totally failed to look at other factors too, i was bearish on in for 2 days, and after i saw the rate...
Well I don't, I speculate At the end will see how good is my speculation

https://charts.mql5.com/5/288/usdchf...ari-uk-ltd.png

https://charts.mql5.com/5/288/eurjpy...ari-uk-ltd.png
0 traders viewing now