Higher Edge within a Single Candlestick | Page 58 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

An Edge for an Edge (System Development) 116 replies

Beating the Odds - Edge Within a Single Candlestick (Explained) 71 replies

Looking for higher timeframe candlestick overlay indicator 10 replies

top 2 single candlestick pattern 7 replies

There is No Edge in a Single Candlestick 64 replies

Attachments: Higher Edge within a Single Candlestick
Exit Attachments

Higher Edge within a Single Candlestick

  • #1,141
  • Oct 5, 2017 10:02pm Oct 5, 2017 10:02pm
Guys, nobody is arguing that the probability of getting either heads or tails is different than 50% in a coin toss. Not even CrucialPoint is arguing it. What he is saying is that if you flip a coin ONCE, there can only be two combinations of results-- Heads, or Tails. 1H/0T or 1T/0H

If you flip the coin 500 times in a set, rather than once, the probability (percentage of the time) that you get exactly 250 heads and 250 tails[50/50]) is very low. Take 10 sets of 500 flips. The probability of getting 250 heads and 250 tails (50% probability) is very low for each set. Rather, you can bet that it WON'T be exactly 50%.

Let's say you have two bets. Either you bet it will be exactly 50% (250 heads and 250 tails) or NOT 50% (ANY COMBINATION AS LONG AS ITS NOT 250/250)

If its a bet on ONE coin toss, then you have no edge. The chances are 50/50.

If it's a bet on 100 coin tosses, you can bet that the result probably will not be a perfect 50% heads, 50% tails.

Everybody will bet the probability will not be 50%.

Think outside the box!
  • #1,142
  • Edited 10:23pm Oct 5, 2017 10:12pm | Edited 10:23pm
Quoting Branflakes
Guys, nobody is arguing that the probability of getting either heads or tails is different than 50% in a coin toss. Not even CrucialPoint is arguing it. What he is saying is that if you flip a coin ONCE, there can only be two combinations of results-- Heads, or Tails. 1H/0T or 1T/0H If you flip the coin 500 times in a set, rather than once, the probability (percentage of the time) that you get exactly 250 heads and 250 tails[50/50]) is very low. Take 10 sets of 500 flips. The probability of getting 250 heads and 250 tails (50% probability) is very...
Oh yes it will. Even if its not %50 In long term it will be around just %53-54 which is still not enough to make profit due to spreads.
If you don't have edge dont even waste time to breaking some hidden secrets here.

Markets are based much more complex facts. If you understand the complexity you can explain them in simple terms..

Trade ticks, days , Hour charts doesnt matter at all. Its all about what you calculate , What you looking at ? Where is your prediction ?

If you don't have data , you cant learn what made profit. Stop wasting time around and trying to use cryptology on some GURU's post. For me they are all clown.

Giving some garbage to make people play with..

Claiming they cracked the market by doing simple volatility and physics calculations on tick charts. Move on please.

If you would give the similar effort to understand what jim simons say on ted talks you would already have more understanding about how algo trading works. In fact most of people not even heard about him. Its like trying to be at the top %5 inside of football players but have no clue about who messi is.

I don't say the knowladge here is useless but look how many person made up something with it ? :d Even 20$ paid sup dem course will give more benefit than all this garbage and self ego feeding..

I am pretty sure on the background real profit makers just laughing to dead and rolling on the floor because of this comics.

Quants out.
Take it easy.
  • #1,143
  • Oct 5, 2017 10:22pm Oct 5, 2017 10:22pm
Quoting Quants
{quote} Oh yes it will. Even if its not %50 In long term it will be around just %53-54 which is still not enough to make profit due to spreads.

Sigh, you're still not reading between the lines. You, again, focused on one set. If you have one instance of throwing coin tosses, the first toss will either be a heads or tails. Nobody is doubting that. THAT is a perfect 50/50. However, when will you reach this perfect 50/50 again? At 100 coin tosses, the odds of it either being

1) 50/50
2) Anything except for 50/50

will favor (2). Over the course, for example's sake, 50,000 coin tosses, the percentage will indeed reach near 50/50.

But let's look at 100 sets of 50,000 coin tosses.

Set 1 will be 24,900 heads, 25,100 tails.
Set 2 will be 27000 heads, 23000 tails.
Set 3 will be 24537 heads, 25463 tails.
SET 4 WILL BE EXACTLY 25000 heads, 25000 tails.
Set 5 will be 26452 heads, 23548 tails.
And so on...

Do you see? The chances of you NOT having exactly 50/50% odds are WAY HIGHER than your odds of having exactly 25000 heads and 25000 tails.

Don't limit yourself to one perspective. Nobody said you have to play one singular set and to make that set impossibly long.

I'm not advocating any strategy, I'm just clearing up any misunderstanding about CrucialPoint thinking the odds on ONE single coin toss are not 50/50.
  • #1,144
  • Oct 5, 2017 10:26pm Oct 5, 2017 10:26pm
Quoting Branflakes
{quote} Sigh, you're still not reading between the lines. You, again, focused on one set. If you have one instance of throwing coin tosses, the first toss will either be a heads or tails. Nobody is doubting that. THAT is a perfect 50/50. However, when will you reach this perfect 50/50 again? At 100 coin tosses, the odds of it either being 1) 50/50 2) Anything except for 50/50 will favor (2). Over the course, for example's sake, 50,000 coin tosses, the percentage will indeed reach near 50/50. But let's look at 100 sets of 50,000 coin tosses. Set...

You have only 1 account to trade , Not set of account.
Even after when you average the sets you still have %50 more or less probability to be profitable or not. Which will be blocked by spreads and comissions again.
Some deep loosers or deep winners will cover other sets. At the end No edge. It can't be predicted unless there is a pattern.

Without research in sample and out of sample data you cant learn whether that pattern worked.

Its like picking 10x Random strategy and hoping at the end being profitable.

Beginning of statistics when they speak about game theory teacher said

Always tails come. Theory says you can throw coin 20 times and 20 times in a row you can have HEAD.

Maybe %0000000000.1 probability you can have 5000 tail streak. But heads always COME ! wanna try ? This is the only edge you can get from all this garbage.

This can be the edge that you know in some streak there will be heads coming. Derren brown was funny actually. Look what he did here after how many hours of attempt ?



Inserted Video

Inserted Video
Take it easy.
  • #1,145
  • Oct 5, 2017 10:28pm Oct 5, 2017 10:28pm
Quoting Quants
{quote} You have only 1 account to trade , Not set of account. Even after when you average the sets you still have %50 probability to be profitable or not. Its like picking 10x Random strategy and hoping at the end being profitable.
You have ONE account, but you don't place ONE trade. You can place 100 trades a day, 1000 if you're willing.

Why limit yourself to one set (one trade in my above example)

When you can have any number you wish.


Again, not my strategy. I'm just clearing up some ways of thinking about probability.
In one toss, chances of it landing on either end is 50/50.
For every single instance that will be true.
However, if you look at the end result of a multiple number of tosses, you can bet that it won't be exactly (total toss/2) heads and (total toss/2) tails.
  • #1,146
  • Oct 5, 2017 10:34pm Oct 5, 2017 10:34pm
Quoting Branflakes
{quote} You have ONE account, but you don't place ONE trade. You can place 100 trades a day, 1000 if you're willing. Why limit yourself to one set (one trade in my above example) When you can have any number you wish.
Bullshit. Place 10.000 trades
This is still clueless trading. You can't predict the drawdown.
Still based on luck factor whether it will go profit or not for 1 time. If you cut trading while you are profitable you have a chance.

There is a word. "If you have unlimited amount of money and unlimited amount of time. Unlimited amount of trade placing" Winning is guaranteed.

Find me a broker that will let you trade more than 120 trades at once.

Only broker you can set 1000 trades is OANDA and they have FIFO Rule which means you cant close the rest without closing the first one.

Oh there is FXPIG aswell.

But anyway.. Keep faith ! Lets trade like a coin toss and people like jim simons will invest on scientists , mathematicians , strong computing power and strong data analysis with brilliant minds in a room of people.... But fk all this shit!!

All we need some CP POSTS with some pictures and bluesharks in it !
Take it easy.
  • #1,147
  • Oct 5, 2017 10:38pm Oct 5, 2017 10:38pm
Quoting Quants
{quote} Bullshit. Place 10.000 trades This is still clueless trading. You can't predict the drawdown. Still based on luck factor whether it will go profit or not for 1 time. If you cut trading while you are profitable you have a chance. There is a word. "If you have unlimited amount of money and unlimited amount of time. Unlimited amount of trade placing" Winning is guaranteed. Find me a broker that will let you trade more than 120 trades at once. Only broker you can set 1000 trades is OANDA and they have FIFO Rule which means you cant close the...
Where did I write the words "At once."? Where did I imply you had to make 120 trades at once?

Do you toss 50,000 coins at once? No. You go one after another. Jesus.

I'd bet my behind off that any broker will let you place a total of 120 trades a day. Heck, why not 1,000.
  • #1,148
  • Oct 5, 2017 10:44pm Oct 5, 2017 10:44pm
Quoting Branflakes
{quote} Where did I write the words "At once."? Where did I imply you had to make 120 trades at once? Do you toss 50,000 coins at once? No. You go one after another. Jesus. I'd bet my behind off that any broker will let you place a total of 120 trades a day. Heck, why not 1,000.
If you place trade one after one , Then its 1 account. 1 set. Still you can't come with an edge. There is no edge within a single candlestick. No one can prove it.
Most of brokers will let you keep only 100 trades at one time. Oanda is the only one goes for 1000.

Its all garbage. It reminds me merlins AMAZING STRATEGY

If you want an edge , Take a look last 10.000 candlesticks and ask them question. Then you will have a probability for 10.001 th One but still it wont be guaranteed enough and there are many factors are missing.

Lets take the edge of 1 candle stick.

It did close last 1000 mins MORE UP Candles than DN ones.
You have 440 mins left. Trade it. Trade only BUY. .. Its all you can get, This is all best edge you can get . And still , If you did not tested it on enough candlesticks.. It will be still pure gambling.

You can't find the answer without looking previous examples. Especially if you are following a rule.
Take it easy.
  • #1,149
  • Oct 5, 2017 10:49pm Oct 5, 2017 10:49pm
Quoting Quants
{quote} If you place trade one after one , Then its 1 account. 1 set. Still you can't come with an edge. There is no edge within a single candlestick. No one can prove it. Its all garbage. It reminds me merlins AMAZING STRATEGY
Okay big boy, you win
I'd bet you never seen Jupiter with your own eyes. But does it mean there is no Jupiter?

History has taught us that people who make blanket statements are the most ignorant of all.
"We can't ever go from China to US in less than a day"
"We can't ever reach the moon"
"We can't ever blah blah"
Just because you can't see an edge, does not equate to "There is no edge."

Now instead of quoting CP, here, I copy and pasted what he said.

""

When you flip a coin, the outcome is either head or tail. What is the probability of flipping a head? What is the probability of flipping a tail? These are not a trick question, and the answer is 50% probability.
This is the level of understanding of everyone. This understanding is right. I have nothing to add here.... but this does not apply to trading. And people are stuck at this level.

So far, so good? Let's take this to the next level.

The probability of flipping either a head or a tail is 50%... If you only do this once!!!!
I can honestly and confidently say, that if we flip a coin 2,3,4-10x, I wouldn't be able to predict the outcome. Nor you or anyone else. In fact, if we surveyed everyone's prediction, we would get a bell-curve statistical result.

If we repeat this experiment several hundreds, thousands, million of times, we can get 10 heads in a row, 10 tails in a row, 7H and 3T, 4H and 6T, etc... Just like Pi, we would get an infinite number of combination of Heads and Tails, which NOWHERE will come close to a 50% probability. When you flip a coin ONCE, you will get a 50% probability... When you flip a coin for INFINITY, the probability is NULL.

""

Now I do admit his wording is different than my wording, my example is the same as his, except for the substitution of some words that might be confusing to some and which would imply different things.

However look at my posts above, and see if you can draw a correlation from my examples to what he wrote.
  • #1,150
  • Oct 5, 2017 10:52pm Oct 5, 2017 10:52pm
This garbage speaking wont lead anyone to success for trading.
Instead of wasting 30mins , 1hour , 4hours a day to cracking this bullshit. Start looking for real examples what real people did. Learning maths , macroeconomics. Data analysis , Programming.

This is just the place that you can laugh or fish some newbies for some green.
Take it easy.
  • #1,151
  • Oct 5, 2017 10:58pm Oct 5, 2017 10:58pm
Quoting kprsa
{quote} We just showed that the long sequence of n random steps, if it is to end above the reference line will look like a “up” candlestick with an open value of 0 (ref line), close value of sqrt(2 n /pi) and an expected high value of sqrt( pi n/2)... So I am actually quite convinced it is beatable - it wouldn’t be in case of a doji i.e., if we wouldn’t buy above the line and sell below it. CP is right imho. k
This is a very good post of you kprsa. This remind me on a LinK some where (unfortunately, i can not find it at the moment...) that there is a foot print on the candlestick, proving that market is not random. If i print a candlestick chart Above a piece of paper (lets say TF M1 of EURUSD chart), i still can not prove that it is not random. I choose EURUSD as my testing chart, because it is (almost always) in Balance between the UP/DN Strength, some times we can find this by looking at the Wick of the candle (Upper/Lower Wick), based on its Open/Close price. If i also print the candle of M5 or H1 chart (i just try to see from small to Big TF) for example, i still get the same Problem, nothing is special there. But this Time Bar has some thinK interesting if i draw some lines on the printed chart, lets say i draw the horizontal and vertical lines of the high & low of the Time Bar. I find there is some Similarity there, but not so sure (i can be wrong though). Any idea, what must we do, after we found like this Problem ? This candle stick can not tell us the direction ... , or i am totally wrong ? Any way, i still believe there is gold in here that can be found .... Gracias !!
I am failed,but i found 1,000 ways that won't work ...
  • #1,152
  • Oct 6, 2017 12:01am Oct 6, 2017 12:01am
Quoting Quants
{quote} Oh yes it will. Even if its not %50 In long term it will be around just %53-54 which is still not enough to make profit due to spreads. If you don't have edge dont even waste time to breaking some hidden secrets here. Markets are based much more complex facts. If you understand the complexity you can explain them in simple terms.. Trade ticks, days , Hour charts doesnt matter at all. Its all about what you calculate , What you looking at ? Where is your prediction ? If you don't have data , you cant learn what made profit. Stop wasting time...

The future depends on what we do in the present.
  • #1,153
  • Oct 6, 2017 2:42am Oct 6, 2017 2:42am
Probably its me...BUT looks like guys here have now resorted to "Who can write up the most complicated stuff...that people won't understand!!"
We all know price is always 1. Moving up Or 2. Moving Down. Or 3. Not moving at all.
___
As if ..Like trying to solve trajectories on how to kill a mosquito with an anti Aircraft missile!!
L.L.L.T
  • #1,154
  • Oct 6, 2017 2:52am Oct 6, 2017 2:52am
Quoting KayStreet
Probably its me...BUT looks like guys here have now resorted to "Who can write up the most complicated stuff...that people won't understand!!" We all know price is always 1. Moving up Or 2. Moving Down. Or 3. Not moving at all. ___ As if ..Like trying to solve trajectories on how to kill a mosquito with an anti Aircraft missile!!

Attached Image
  • #1,155
  • Oct 6, 2017 3:11am Oct 6, 2017 3:11am
Quoting Copernicus
{quote} {image}
To be more precise it's denial of the PDM category.
  • #1,156
  • Oct 6, 2017 3:27am Oct 6, 2017 3:27am
its pretty simple , if you're asked to toss a coin ,heads or tails , make sure whatever YOU want to come up is on top IF you catch coin in your hand and flip it over ... try it and see
  • #1,157
  • Oct 6, 2017 4:05am Oct 6, 2017 4:05am
Quoting mzvega
{quote} No it's a way of measuring demand…….. https://www.forexfactory.com/showthr...48#post9022648 I need some time to put my response in context………..
The reason AMVT uses Market generated info/data (price and the time in which it occurred) is that you need both price & time to measure demand..

Market generated info/data includes the element of time. As I stated in my previous post the nature of OHLC excludes the element of time. Not a permitted practice to use for analysis, You need more info than just Price alone.

No matter how one tries measuring data linearly (bar chart/candlesticks) "price against price". Any measure of price linearly from Point A to Point B can never measure demand (price*time = value/demand) The data is just not there.

OHLC is a linear measure from high to low or low to high it is a linear 1 dimensional series of data

Price* time = value (demand) it is a 2dimensional measure

Measuring one price against another price is a 1 dimensional linear measure

You can't solve a 2dimensional question (price*time= value) using a 1 dimensional series of data (OHLC)

Measuring one price against another price is a 1 dimensional linear measure
Measuring "price over time" is a 2 dimensional measure.

Example"
You need 2 variables , Height*Diameter to solve the volume equation.
You need 2 variables Price * Time to solve (measure) price*time = value/demand
Attached Image


It is impossible to solve the Volume equation if the only data (variable) you have is height.
It works the same way with price, you can't measure demand (price*time = value/demand)
equation when all your working with is " Price" ( linear price to price OHLC data ).. The data is just not there.

When the only measure (variable) you have is height, you cant accurately measure the volume, it's impossible. Height is a Linear measure. As is measuring one price against another price is a Linear measure. When Price is the only measure (variable), your working with you can't measure price*time = value/demand
The data is just not there.

Linear measures using just price is the same as trying to measure the volume of an object, when the only info you're working with is height..

You have to ask yourself if you are using the right data for the questions being asked?.............
Or do you expecting way too much from you data?……..
Attached Image
Markets are not efficient, rather they are effective - Jones
  • #1,158
  • Oct 6, 2017 4:06am Oct 6, 2017 4:06am
Quoting Branflakes
{quote} Sigh, you're still not reading between the lines. You, again, focused on one set. If you have one instance of throwing coin tosses, the first toss will either be a heads or tails. Nobody is doubting that. THAT is a perfect 50/50. However, when will you reach this perfect 50/50 again? At 100 coin tosses, the odds of it either being 1) 50/50 2) Anything except for 50/50 will favor (2). Over the course, for example's sake, 50,000 coin tosses, the percentage will indeed reach near 50/50. But let's look at 100 sets of 50,000 coin tosses. Set...
I dont believe coin tossing can be compared to trading, so I don't really see the point in doing so.

But for the sake of argument, I think it can be proven quite easily that coin tossing should not be compared to markets,
by the following experiment.

1. Lets take a 1000 cointosses (e.g. per hour).
2. we start with 0 tosses, and the reference count = 0 (= Open)
2. count the number of heads and tails, where heads = +1 and tails = -1, sum them together cumulatively.
3. after this period of 1 hour, determine the highest and lowest points reached (= High and Low of the 1H period).
4. determine the final sum of heads and tails (= Close of the 1H period).

repeat this experiment at will. Basically it will result in a series of OHLC candles (where Open = previous Close).

Now the question is, what will be the distribution of the wicks and body of these candles?
will it be 25-50-25 as in markets candles, or will it be 33-33-33 as in True random candles?
  • #1,159
  • Oct 6, 2017 4:28am Oct 6, 2017 4:28am
Quoting despacito7
{quote} Ththat there is a foot print on the candlestick, proving that market is not random. If i print a candlestick chart Above a piece of paper (lets say TF M1 of EURUSD chart), i still can not prove that it is not random. I choose EURUSD as my testing chart, because it is (almost always) in Balance between the UP/DN Strength, some times we can find this by looking at the Wick of the candle (Upper/Lower Wick), based on its Open/Close price. If i also print the candle of M5 or H1 chart (i just try to see from small to Big TF) for example, i still...
If I understand you correctly: I agree with you, there is a self-similarity across timeframes (I refer to it as “music”) I agree also that this is an evidence of a non-random market. Probably what happens is that the big market players test the market’s susceptibility to smaller “shocks” before embarking on a bigger and longer push. I use this in my trading for various estimates. There was a member called skunny that outlined a full trading method based on the same principle.
k
  • #1,160
  • Oct 6, 2017 5:05am Oct 6, 2017 5:05am
Quoting strongtrader
{quote} So, correct me if i?m wrong, your point is that the tick frequency over a given period in time tells you price direction. In other words this is momentum; but momentum is not a constant, so it can give clues about direction that are very short living as momentum can vary very fast. A shift in momentum does not mean necessarily a change in direction. For sure you can trade momentum but it is just a way between many others. The key is first to understand "the why of things", thereafter you will be able to do whatever you want getting consistent...
AMVT trades change in demand/value
TA trades or follows changes price…….

Both come from 2 very different "schools of thought"

From an auction market theory perspective Price is nothing more than an advertizing mechanism. It's its part of the "price discovery process". The Price/ticker/quote is advertizing. It moves to solicit and encourage trade (looking for a response) It moves in search of participants in order to facilitate trade. Different prices at different times attract different TF participants. Price auctions too high in search of sellers and auctions too low in search of buyers. If no orders are placed price will definitely move. That is the purpose of auction markets, to facilitate trade. If no orders are placed price will auction up & down (price discovery) looking for new participants or move to an attractive enough price to get the participants already holding trades to change their position.

Price discovery is what gets a specific buyer and specific seller to move from establishing a general price to agreeing on a specific price for their transaction, based on the size of the transaction, location of the transaction, cost of the transaction, and many other factors. It is a dynamic process, and in a way, it is the true mission of any market and any exchange.

Price discovery is one of the central functions of financial markets. In the market microstructure literature, it has been variously interpreted as, the search for an equilibrium price (Schreiber and Schwartz (1986))

Price moves because it is a function of the Auction Process……...

From an auction market theory perspective I have no interest in the advertizing mechanism, it's about measuring/trading "Demand". Knowing how and where prices are being used over time, give you a quantifiable measure of demand. Price fluctuates that’s a given,

Decisions to buy or sell are not made on the basis of the "direction" of price, in which you imply. If there is an imbalance in favor of prices being used by sellers than buyers, then depending on the condition of the market, one would also consider being a seller……..until either something in the condition of the market changes, and or there's a change in the imbalance.

It doesn't determine the direction of price, It's the measure of demand I am concerned with, it tells you that if there is an imbalance in favor of prices being used by sellers than buyers, one would trade the odds in favor of the imbalance, one would be looking to be a seller.

It's not used to determined the direction of price, per se, it's used to determine the odds of which side of the market you ought to be on…….


So, back to the original question
Quoting strongtrader
So, correct me if I'm wrong, your point is that the tick frequency over a given period in time tells you price direction.
Not in the "linear" sense in which you imply………..
Markets are not efficient, rather they are effective - Jones
0 traders viewing now