Edge in an automated entry? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

An Edge for an Edge (System Development) 116 replies

Scripts for Limit entry, Stop entry, Market entry 47 replies

Automated or semi automated trading? 3 replies

Manually Drawn Zones, Automated Entry and MM 8 replies

Automated order entry at selected time, to play daily high/low ... 0 replies

Edge in an automated entry?

  • #1
  • First Post: Dec 19, 2011 10:19am Dec 19, 2011 10:19am
How does one find if a rule based entry methodology has an edge? Is it even possible for an entry to have an edge when other aspects of the system are completely random? If so, how can one establish this statistically and without curve fitting?

Any suggestions of such a rule based entry are welcome.
  • #2
  • Dec 19, 2011 10:46am Dec 19, 2011 10:46am
Ideally they'd be some logic behind the rule based on market structure to back up your stats but either way, for it to be relevant I'd want to see a defined edge based on 1:1 risk:reward. You'd test over a range of values, and it should be pretty conclusive over all of them. Otherwise, chances are, you're dabbling with randomness.

You wouldn't necessarily trade 1:1 of course.

You'll quickly realise nearly everything that's good isn't public, for obvious reasons.
  • #3
  • Dec 19, 2011 11:09am Dec 19, 2011 11:09am
Quoting triphop
Ideally they'd be some logic behind the rule based on market structure to back up your stats but either way, for it to be relevant I'd want to see a defined edge based on 1:1 risk:reward. You'd test over a range of values, and it should be pretty conclusive over all of them. Otherwise, chances are, you're dabbling with randomness.

You wouldn't necessarily trade 1:1 of course.

You'll quickly realise nearly everything that's good isn't public, for obvious reasons.
I guess we should wait for other responses but I think, universally speaking, an entry can never have an edge. I will be happy if I am proven wrong.

The reason I am saying so is this. A universal edge in the market is follow trends, minimize risk/maximize reward by responding to the market which cannot be predefined at the time of entry. Just by an entry, one cannot make use of these market inefficiencies and therefore entries cannot have an edge.

Having said that, in certain cases entries can have an edge, but those cases will be curve fitted and will not apply universally to all markets and will therefore fail in due course of time as markets change.
  • #4
  • Dec 19, 2011 11:41am Dec 19, 2011 11:41am
You've just set yourself a big obstacle in my eyes and I'm not going to prove you wrong; that's for you to figure out off the back of a few thousands hours of testing.
  • #5
  • Dec 19, 2011 1:19pm Dec 19, 2011 1:19pm
Isn't your logic upside down? You'd pick an entry point because you've determined that statistically it would result in a profit /given your parameters, possie size, SL, TP, and such/

Yes your random entry can certainly have edge if you have friends in Frankfurt ... at the ECB

Quoting Eklavya
How does one find if a rule based entry methodology has an edge? Is it even possible for an entry to have an edge when other aspects of the system are completely random? If so, how can one establish this statistically and without curve fitting?

Any suggestions of such a rule based entry are welcome.
forget about tomorrow, just steal away into the night
  • #6
  • Dec 19, 2011 1:28pm Dec 19, 2011 1:28pm
Quoting spekitox
Isn't your logic upside down? You'd pick an entry point because you've determined that statistically it would result in a profit /given your parameters, possie size, SL, TP, and such/

Yes your random entry can certainly have edge if you have friends in Frankfurt ... at the ECB
So you are saying that entry can have an edge? In other words, it is possible that a system that employs such an entry and has other parameters random (say 1:1 RR ratio) is profitable? If so, can you outline your rationale please?

If this does not make sense, how would you present your argument in favor of this statement. "It is possible for an entry to have an edge."

There is no debate that systems having a logical entry, exit, money management, trade management etc are profitable. My point is focused on entry only. Can Entry only ever give you an edge or you always need the other ingredients to make a system profitable.
  • #7
  • Dec 19, 2011 3:30pm Dec 19, 2011 3:30pm
No bro, I'm not saying entry can have an edge. I think entry itself, is just Mickey mouse. It's the exit that will give you a profit /or a loss.

But in a way, your question reminds me the old rabbi joke. Two youngsters ask him if it's allowed to smoke while learning Torah. No, strictly forbidden! - so is the answer. Then they ask it the other way around - is it allowed to learn Torah while we smoke? Yes, absolutely - the rabbi said.

So, what is your question? Is it possible to find a profitable set of parameters for about any kind of entry signal ever invented by man? Yes, if you have enough time, you will find the perfect settings to build a profitable system around ANY entry signal.

Or, is it possible that the "perfect entry" will keep making money no matter how you fiddle with the system parameters? No, I'm afraid that's just another way to get the margin call.

Quoting Eklavya
So you are saying that entry can have an edge? In other words, it is possible that a system that employs such an entry and has other parameters random (say 1:1 RR ratio) is profitable? If so, can you outline your rationale please?

If this does not make sense, how would you present your argument in favor of this statement. "It is possible for an entry to have an edge."

There is no debate that systems having a logical entry, exit, money management, trade management etc are profitable. My point is focused on entry only. Can Entry only ever...
forget about tomorrow, just steal away into the night
  • #8
  • Dec 19, 2011 5:57pm Dec 19, 2011 5:57pm
Quoting Eklavya
How does one find if a rule based entry methodology has an edge?
I think only by split-testing it against random entries, duplicating all other randomised aspects of the trade, in each case, over a statistically signifcant number of theoretical trades.

It must be possible (albeit not easy) to do this.

It's quite an interesting concept. At least, it is to me, because I happen to believe that "entry alone" isn't all that important a component of a profitable trading strategy, but at the same time it's very clear that some entries are better than others!
  • #9
  • Dec 20, 2011 1:25am Dec 20, 2011 1:25am
Quoting Eklavya
How does one find if a rule based entry methodology has an edge? Is it even possible for an entry to have an edge when other aspects of the system are completely random? If so, how can one establish this statistically and without curve fitting?

Any suggestions of such a rule based entry are welcome.

Hi, just stumbled across your thread.

There is a way to have an edge and is used everyday by the professionals, but it includes a lot of mathematics and therefore computers in order to take advantage before the edge disappears.

What is it you may well ask and the answer is SKEW.

As you said yourself assume that the market is random then any price distribution should plot a standard bell curve but, markets don't always do this and then the distribution model gets skewed producing opportunities with an edge...

The maths is complicated but this is one of the most commonly used techniques for spotting market opportunity with an edge as used by all the commercial and professional players.

There will be plenty of internet material on price distributions and skew but it won't be light reading enjoy.
  • #10
  • Last Post: Edited 7:27pm Dec 31, 2011 4:27pm | Edited 7:27pm
For starters, exactly what constitutes "entry" needs to be defined.

If (as a theoretical example) a certain S/R setup allows me to place a very tight SL that (on the basis of historical probability) delivers a high RR trade and consequent overall positive expectancy, does 'credit' belong to the entry, or the way in which I manage the exit?

For me, then, the argument over whether entries or exits are more important has become somewhat meaningless (rather like debating whether offense or defense is more important in a game of football). The key is to use the two in complementary fashion to establish a profitable method.
0 traders viewing now