Hi
Just a quick though , I know the saying "let winners run" etc is good advise but what is the average target for points ie a 10 point target for a 5 point stop on a day trade or wider for long term positions .
More curious than anything else to see if small wins out weigh the big wins as I seem to see the losses rack quicker when looking for the big push
This is where the yodaesque trading advice of "develop a trading system that suits your trading style and personality" comes into play in my opinion. I always hated reading that advice since we all have the same goal...make more than we lose...so really how can anyone have a "style" of trading right?
Really one main thing (in my opinion) that it comes down to is your tolerance to losses...how many in a row can you have before you lose confidence in your system and walk away from it? In general when developing a system with targets, the smaller your profit target is with respect to your stop loss, the higher probability you have of a positive outcome with the trade and vice-versa.
From my reading/research, most books and articles that talk about system development focus more on developing, and in general seem to overwhelmingly recommend the development of, systems with higher profit targets than stop losses. Reason being, as long as your targets are higher than your losses, you don't need to have a majority of winners to be profitable. A reasonable argument for sure and sound pretty good to me, BUT the one thing that the vast majority of books and articles overlook is how to know when to stop trading a system and how to determine the system even works in the first place. If you are trading a system with a low probability of winning, I posit that it is very difficult to know when it has gone ape ****. With a high probability win system, however, you will know very quickly when it goes ape **** and/or if it even works.
Taking a closer look at this point: the probability of x number of losses in a row when the total trades in the system reaches y can be estimated using a binomial distribution. As an example, if I expect to make 25 trades over a certain timeframe the number of losses I am likely to see would be as follows for various percent win systems:
From this it should be clear that with a higher percent win system you will know faster and with greater certainty if something is wrong in the system. Taking it to the extreme of a system with an only 25% win rate, there is a 50% chance I will have 19 losses in any 25 consecutive trades I take and I am almost guaranteed 15 of those trades will be losses. Furthermore, there is still a finite (>1%) chance that 23 of my 25 trades will be losses. To put that in perspective, that is at minimum 8 losses in a row and very likely more than that.
Since the probability of a system will fluctuate some over the course of trading it, how can I possibly know with any certainty my system has gone wrong if I designed it with only a 30% win rate? It is reasonable to assume a +/- 5% fluctuation from small set to small set of trades so a 25% win rate is within the realm of possibilities of such a system even though it may be functioning correctly. I can only really start to suspect something is wrong with it after the 24th consecutive loss.
The person who can deal with a system like that would either need a very long proven track record of the system working, or needs to have a very high tolerance to losses. The position sizes that would be possible with a system like this would likely be quite low since the drawdown from 24 straight losses would very likely be astronomical otherwise (even with a very high profit target to stop loss ratio).
Another thing that pretty much every book and article overlooks is that a profit target/stop loss system is not actually simply a binary system. There are 4 possible outcomes: reaching profit target, not reaching profit target but still having a profitable trade, not reaching stop loss but still having a losing trade, reaching stop loss. It could be argued that a 5th outcome is possible - profit above profit target (also could be argued that loss greater than stop loss is possible but I will make the assumption that you are following your system and there is no oddity/black swan event involved in the trade).
So even if your system is developed so that, in theory, you have a stop loss 5 times greater than your profit target, if your stop loss is only tripped 5% of the time, the system won’t be nearly as bad as it appears to be at first glance. Sure there is a potential for a bad day to wipe away weeks worth of profits, but if this only occurs once a year it won’t be terribly difficult to take as long as you always keep in the back of your mind that there is the potential for this to occur.
Really the question becomes, how long are you willing to wait for your developed system to prove to you that it works, and how far are you willing to trade it into losses before it proves to you it has stopped working? I know for me, when I have money on the line (which is really the only way to prove a system), the answer is a resounding NOT VERY LONG! I am a fairly patient person, so I would conjecture the majority of people would answer the same.
Having said all of this, the margin for error on a system that has a lower profit target than stop loss is generally razor thin. With smaller strings of losers though it is much easier to gain confidence in your system and be able to follow it unquestionably which is what needs to be done for a system to function properly. What is the use of rules if they can't be followed?
. I know (more or less) that if I achieve 1.5 to 1 my account will tick along nicely, so that's what I want out of a trade when it goes right and anything more is a bonus.
