How to avoid the bad slippage? - page 2

New comment
 

Thank you Malacarne for your sharing. I agree "no execution" is better than a bad execution with high slippage. In reality with instant execution I have received thousands of "re quote issue" when the market is busy .. Even with 1 pip slippage in OrderSend method.  

 
Could it be broker issue. I just read one thread on FF. Someone trading 500 lots size got no problems. He says must use stp or ecn broker
 
Anton Nel:

Thank you Malacarne for your sharing. I agree "no execution" is better than a bad execution with high slippage. In reality with instant execution I have received thousands of "re quote issue" when the market is busy .. Even with 1 pip slippage in OrderSend method.  

1) Don't trade when the markets are too volatile (e.g. some news time).

2) Change your broker!

 

My currently best EA performs extremely well in tester with 200-400 trades per day but is still work in progress. The final problem to solve for it is the slippage in stop prices. My Broker told me the following.

  1. "There is no slippage in testing, at all." Thus when after opening a position at price o the price moves to price c beyond the stop price s that the EA requested then the position is closed with the stop price s. Since the EA is programmed to trail the stop prices as profit increases and always computes s such that a loss cannot exceed a suitable percentage of the balance this is necessarily quite profitable.
  2. "In live trading positions can only be closed at available prices." This certainly makes sense. Why should my broker let me close a position at price s if the price has already moved to c which is worse for me?

My conclusion is that slippage is not a fixed number of points my broker tries to take away from my profit in every trade. A broker who is the source for slippage by taking a few points from every trade should not even be able to compete with other brokers for a long time. Thus it is not necessarily right to put the blame on the broker. Preferring "no execution to bad execution" seems like a triviality to me. However, when prices move beyond stop prices it is mostly even worse not to close a position because the current price is already more pips away from the desired stop price than the number of points we are allowing in the slippage field of the order request stucture.

  • If my broker's "available prices" are those that are passed to the OnTick() handler too using SL prices in orders should be equivalent to checking for triggered stops and potentially closing positions on every tick. I hope that this manual emulation of SL'es would give me the same results in tester as in live trading. Can anybody confirm this?
  • Has anybody a universal solution for such a manual emulation of stop-losses?

 

 
Thomas Schwabhäuser:

...

  • If my broker's "available prices" are those that are passed to the OnTick() handler too using SL prices in orders should be equivalent to checking for triggered stops and potentially closing positions on every tick. I hope that this manual emulation of SL'es would give me the same results in tester as in live trading. Can anybody confirm this?
  • Has anybody a universal solution for such a manual emulation of stop-losses?

 

Slippage is a part of the "game", you can't avoid it.

What you can avoid is broker bad practice, by choosing one which doesn't manipulate prices.

 
If a broker new or they dont have enough offers from LP's their slippage will be high when broker have good LP's circles they can fill your level with less distance.  
 

Thanks to Alain and Waseem for your help to refine my perception of the "game". As trading and gambling do have some common traits the following 2 links may be an interesting complement:

  1. https://en.wiki2.org/wiki/Bookmaker
  2. https://en.wiki2.org/wiki/Mathematics_of_bookmaking
However, I still believe that fast price movements play a more significant role when positions are not closed at their stop prices than the slippage added by the broker does. At least, when I am on the M1 chart of a not so quiet market. I have just confirmed that applying the same EA to the M5 chart makes it more survive longer, M10 even longer, but M20 for a shorter period of time. However, it takes H4 to make this breakout strategy profitable on 4/21/. Although it is likely to lose on smaller time frames it does not always lose more the smaller the time frame.

I am curious if other traders confirm that slippage is less significant the bigger time frame we choose?

Mathematics of bookmaking - Wikipedia Republished // WIKI 2
Mathematics of bookmaking - Wikipedia Republished // WIKI 2
  • wiki2.org
In gambling parlance, making a book is the practice of laying bets on the various possible outcomes of a single event. The term originates from the practice of recording such wagers in a hard-bound ledger (the 'book') and gives the English language the term bookmaker for the person laying the bets and thus 'making the book'.[1][2] A bookmaker...
 

Actually, slippage does not matter at all, because in the end, the average is zero. Mathematically and statistically. Theoretically. The problem is the broker. I integrated a slippage measuring with my trading system and tested many brokers live in the meanwhile with it. Result: Volatility is not the problem, its just the name of the broker.

Slippage/Requote is exactly the issue where a broker can easily cheat on you, because most people simply do not recognize it and its a bunch of work to track that. There are brokers with low spreads, who call theirselves the most serious and fair, but in the end, exactly these had an average of 1-2 pips requotes in each trade in my tests. I saw the priceline moving over the SL more than 4 pips in quiet markets, just like that, while others have an average slippage which is always equal to zero or slightly positive.

If you "feel" that the execution is bad, then it is bad. Simply change your broker, no matter how good their official reputation is. Dont care about the slippage parameter, set it to infinite but track the results.

And by the way, ECN quoting is good, but the quotes say nothing about the execution. 

 
Idea, may be bad. Nevertheless here my idea: can you avoid the opening of a position with a slippage greater than x by setting the TP x points away from the open price???
Not tested.  Simply an idea. 

Matthias 
 
Dr Matthias Hammelsbeck #:
Idea, may be bad. Nevertheless here my idea: can you avoid the opening of a position with a slippage greater than x by setting the TP x points away from the open price???
Not tested.  Simply an idea. 

Matthias 

No. But you can prevent opening of a position by setting the TP (does not matter how many points) on the wrong side. Example for a Buy position, set TP below open price.

123
New comment