Every indicator can be profitable? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

Is there an indicator that copies the screen at open of every candle? 8 replies

Max and Min price reached in every trade Indicator / EA 4 replies

An indicator to show separation every 4hr 0 replies

Indicator with value on every closing bar 19 replies

Indicator to email price every hour. 4 replies

Every indicator can be profitable?

  • #1
  • First Post: Oct 21, 2013 1:06pm Oct 21, 2013 1:06pm
Do you agree with this article?

"But how to know when to enter or exit a trade? Well, since there is no way to predict future development significantly better than pure guess, you can use anything, any indicator, candle, signal provider, or even to ask your 5 year old daughter: „Hey sweetie, what do you think, will JPY rise or fall today?“ If she says it will rise, you buy it, 1 or 2% risk, SL 100 pips, TP let’s say 300, and after you open it don't touch it until it hits SL or TP, no matter if indicator give opposite signal or daughter changes her mind.

If you are disciplined to do it with no exceptions for 500 or 1000 trades, you will have high probability of making nice profit, because forces that drive markets will be the same, and law of probability will be on your side.

Secret of success is in discipline and RRR, indicators and complicated strategies can be useful, just don’t allow them to distract you from these important concepts!"

Whole article here
  • #2
  • Oct 21, 2013 4:11pm Oct 21, 2013 4:11pm
Total bullshit lol.
  • #3
  • Oct 21, 2013 5:13pm Oct 21, 2013 5:13pm
If you have traded for any decent length of time you will understand how efficient the markets really are and that any indicator used only will be a loss. Through my experience what matters is understanding when a trade is not working and cutting the lose, knowing there is always another trade around the corner, understanding when you have a high probability trade and gearing up, sticking to your trading rules but knowing when to break them and knowing through time spent in the charts, effort and work the probability of success that you have.
  • #4
  • Oct 21, 2013 7:22pm Oct 21, 2013 7:22pm
The guy who wrote the article is both right and wrong.

He is right that discipline is a must have for success, whatever the type of trading.

He is wrong that Risk:Reward Ratio (RRR) alone is the key to success, because:

RRR alone in and by itself is not an edge. It provides no guarantee of positive expectancy whatsoever. RRR doesn't even work alone. RRR is not independent from Hit Ratio (HR%). Specifically, they vary inversely: High RRR (small SL and big TP) = Low HR% (SL is more probable to be hit than TP), and vice-versa.
"To hold, you must first open your hand. Let go." - Lao Tzu
  • #5
  • Oct 21, 2013 8:46pm Oct 21, 2013 8:46pm
Interestingly when you datamine for systems you'll notice that those with out of sample edges tend to have high Risk:Reward ratios. You could therefore say that although a high RRR does not guarantee an edge, it is true that a system with a higher RRR will have a lower chance of being the result of data dredging (in many cases). This is my experience at least. Given two systems with equal in-sample expectancy (RRR*HR%), the one with the higher RRR is more likely to perform well in the future.

Also you can get any indicator and create a logic that gives you any profitability (you can do this through genetic programming as I do). But if you do this without any care you will get to data dredging. Indicators are only mathematical calculations over price and you can use them to find relationships that have historically given excellent trading results. The key is filtering those relationships to find those with a high chance of working in the future (this is the hard hard hard hard part). I would say that any indicator can indeed get you to profitable results, they are all derived from the same information and therefore relationships between them can lead to the same conclusions.
  • #6
  • Oct 21, 2013 10:38pm Oct 21, 2013 10:38pm
This statement is relying on the argument that price is not a completely random motion and have trending characteristics. If you assume that to be true then this works on expected basis but that still doesn't do anything to provide good return to drawdown ratio. Unless the market is heavily trending, you will have very large drawdowns using this method.

Usually there is alpha in momentum signal but that tends to be very weak, so has to be either traded on large number of uncorrelated assets (like Russell 2000 stocks) or enhanced using other signals to get to workable return / drawdown ratio.

My 2 cents
Think : Trade : Live Life
  • #7
  • Oct 22, 2013 7:05am Oct 22, 2013 7:05am
Quoting future
Do you agree with this article? "But how to know when to enter or exit a trade? Well, since there is no way to predict future development significantly better than pure guess, you can use anything, any indicator, candle, signal provider, or even to ask your 5 year old daughter: „Hey sweetie, what do you think, will JPY rise or fall today?“ If she says it will rise, you buy it, 1 or 2% risk, SL 100 pips, TP let’s say 300, and after you open it don't touch it until it hits SL or TP, no matter if indicator give opposite signal or daughter changes...
some people try and reason . some people are just retarded. my bet is on this guy coming from a lottery player perspective. not even proffesional poker player but pure lottery numbers background.
  • #8
  • Oct 22, 2013 2:14pm Oct 22, 2013 2:14pm
This old guy also thinks that future can't be predicted accurately:

http://www.forexfactory.com/news.php?do=news&id=453748
  • #9
  • Oct 22, 2013 3:27pm Oct 22, 2013 3:27pm
Quoting future
This old guy also thinks that future can't be predicted accurately: http://www.forexfactory.com/news.php?do=news&id=453748
LOL the old guy deregulated the market and then he writes a book where he tries to figure it out. Jon Stewart nails him over and over. Love it
"To hold, you must first open your hand. Let go." - Lao Tzu
  • #10
  • Edited 2:08am Oct 23, 2013 12:49am | Edited 2:08am
Quote from the article: "So all you need is positive RRR and to be disciplined to use it all the time, with no exception."

It's easy to disprove this statement: if high RRR and discipline is "all you need", then writing a profitable EA would be child's play, and everybody would run one and become a forex millionaire. Obviously this is impossible in a zero sum game. Markets operate in such a way as to close off easily exploited inefficiencies.

Perhaps ironically, I wish the statement was true, as it would mean that the EAs I've co-developed (here) would be absolutely guaranteed to be profitable. LOL

But as stt points out, high RRR systems are generally dependent on markets 'trending' far enough, often enough.
  • #11
  • Oct 23, 2013 7:10am Oct 23, 2013 7:10am
Quoting hanover
Quote from the article: "So all you need is positive RRR and to be disciplined to use it all the time, with no exception." It's easy to disprove this statement: if high RRR and discipline is "all you need", then writing a profitable EA would be child's play, and everybody would run one and become a forex millionaire. Obviously this is impossible in a zero sum game. Markets operate in such a way as to close off easily exploited inefficiencies. Perhaps ironically, I wish the statement was true, as it would mean that the EAs I've co-developed (here)...
How come I didn't think of that perfect EA example?

The elegant way you think and are able to express yourself never ceases to amaze me.
"To hold, you must first open your hand. Let go." - Lao Tzu
  • #12
  • Oct 23, 2013 3:05pm Oct 23, 2013 3:05pm
Quoting hanover
But as stt points out, high RRR systems are generally dependent on markets 'trending' far enough, often enough.
Good point about EA. Although this guy states multiple time exactly this, he thinks markets do trend just enough.
  • #13
  • Oct 26, 2013 3:20am Oct 26, 2013 3:20am
I agree discipline and RRR contribute to a profitable system. The reason is no one could predict the next movement, if you don't follow rules and trade all the time, you might miss good opportunity. Hence, you might lose over the long term case. For RRR, it has been proven by myself even a lousy system (<50% winning rate) could able to turn into a profitable system. However, return only can be seen for quite a period of time, not in short term case. You may refer to my demo account for this case.

System is important too. Indicators are the tool to assist you to detect or spot scenarios, it doesn't give prediction. Please understand this. Yes, indicator is lagging but you must know how to use them efficiently. Let me share some example.

If you see the candle chart from your eyes, there are many pattern, if you can separate each portion of pattern by using defined indicator. Then it is a good system. This has been proved myself as I just recently doing some hypothesis experiment. One indicator cannot be function well in all scenarios. Just like a robot, it need many sensors to detect each scenarios.
Example for Trending case: 2 scenarios only
a) High jump - Find an indicator to spot this and verify the ending of this jump.
b) Normal trend - The same find an indicator to do same job.

So far I am able to produce around 70% win rate from early 40, 50, 60% of my test. Now the only risk is counter trend.

Just my 2cents.
  • #14
  • Oct 26, 2013 4:30am Oct 26, 2013 4:30am
Quoting future
ask your 5 year old daughter: „Hey sweetie, what do you think, will JPY rise or fall today?“ If she says it will rise, you buy it, 1 or 2% risk, SL 100 pips, TP let’s say 300, and after you open it don't touch it until it hits SL or TP, no matter if indicator give opposite signal or daughter changes her mind. If you are disciplined to do it with no exceptions for 500 or 1000 trades, you will have high probability of making nice profit
Asking a girl her oppinion before she changes her mind is like flipping a coin ROFL. So let's flip a coin and let the little girl play her doll. She isn't interested in markets beyond the colors of the charts anyway :-)

When you totally ignore the market statistics you only rely on the marginal probability of the market going up or down. With FX market this probability is 50% for any move. For SP or Dow better to only buy.

So if I flip a coin and wait either -100 pips or +100 pips, everybody agrees that I win 50% of the time and lose 50% of the time. Either I win 100 or I lose 100. My expectaction is 1/2 * 100 + 1/2 * -100 = 0. A breakeven game + cost of trading.

If I do the same flip a coin game with 100 SL and 200 TP, the market will reach my SL in 1 step 50% of the time. The other 50% I'll have a 100 pips floating profit. I need waiting for the next 100 pips move, either up or down. 50% of the time it will reach my 200 TP and 50% of the time it will return to square 1. So I have 50% losing in one step, 25% winning straight (in 2 steps) and 25% going up then down. The game ends when either the SL or the TP is reached. At each step, as long as the game is running the price is at the entry price. SL has twice much more chance beeing hit than the TP. Trivial equation 2X+X=100. So I have 33.3% chance of winning and 66.6% losing. My expectaction is (1/3) * 2 + (2/3) * -1 = 0. Again a BE game plus cost of trading.

What is really fun is that this is true for ANY value of SL and TP! (I use this model to debug my simulation engine)

If the guy who wrote this article had the discipline he's talking about he would have went to school!!
No greed. No fear. Just maths.
  • #15
  • Oct 26, 2013 6:33am Oct 26, 2013 6:33am
why every one thinks .. flipping a coin .. will give :o 50% win rate ... it can be 30% as well!

you can be wrong 7 times as well ... 7*100 = 700 Pips Loss
and 3 wins = 3*200 = 600 pips gain ...

Lost 100 pips and 2-3 months headech , electricity bill etc etc etc ......
"Know what you own, and know why you own it" — Peter Lynch
  • #16
  • Oct 26, 2013 8:33am Oct 26, 2013 8:33am
Quoting gyanpapi
why every one thinks .. flipping a coin .. will give :o 50% win rate ... it can be 30% as well! you can be wrong 7 times as well ... 7*100 = 700 Pips Loss and 3 wins = 3*200 = 600 pips gain ... Lost 100 pips and 2-3 months headech , electricity bill etc etc etc ......
Should you have properly read my post, the win rate of the second example is 33.3%

If the coin is fair it will eventually give heads 50% of the time. When you toss the coin 10 times you may indeed come up with 7+3. You can and you will also get 3+7 some times. You will get 7+3 as many times as 3+7. The same number of times you'll get 1+9 and 9+1, 2+7 and 7+2....
Probabilities are not calculated from a sample of ten trades.

Sure you will have up trends of several hundreds or pips. But you will have the same number of down trends of the same size at the same frequency. FX market is symetrically distributed around 0. This is why "the flip a coin" (aka random walk) model is a valid simplification.
No greed. No fear. Just maths.
  • #17
  • Oct 26, 2013 12:36pm Oct 26, 2013 12:36pm
To my thinking, yes, as PipMeUp was saying, flipping a coin in the long run (when number of trials tends to infinite) is a 50% heads and 50% tail.

So is the market. Every change in price is either an increase or a decrease. When running the same type of trials as for the coin, there should be something like a 50/50 for up and down.

In addition to the 50/50 chance for direction, there is also another matter, the one of power. This is not a true/false statement, there are multitude of variations in the power of the market participants.

In the end trading with flipping coin is a 50/50 win/loss game, but with unknown R:R every time...
  • #18
  • Last Post: Oct 27, 2013 2:46pm Oct 27, 2013 2:46pm
Quoting PipMeUp
What is really fun is that this is true for ANY value of SL and TP! (I use this model to debug my simulation engine)
How big time period, sample?

Which currency pairs?
0 traders viewing now