Indeed it cannot be used as a risk management. I made a study of the way i trade and the conclusion was that i only had to use a 2 points stop. Therefore this sheet was usefully to me because i always knew where i was standing. If you note down all your trades without using a stop you can calculate where to put the stop to have an optimal result with the best risk/reward ratio and see the maximum drawdown. You can als calculate with more certainty within which limits your results will be in the future, but if you have a good system it should be very clear that in the long term you will never lose money just by watching the trades of your testing.
Many thanks for the constructive comments and criticism in this thread. Nononsense: I've linked directly back to one of your comments from the original article to give readers your perspective which I think is valuable. I've written a follow up called How long will it take to turn $5,000 into $0? which I hope will give more insight into accessing the risk involved and specifically the risk of ruin. One issue that I've been struggling to address is that of being more specific and that has been commented on by a couple of people. Because no particular strategy has been defined here it is very difficult to criticize how money management would be applied to it. The examples that I have give have assumed an edge and a fixed loss and profit figure. I agree that this is not realistic but feel that it's a good starting point.
Stop-loss can be useful in some very special (and rare) cases, but most of the time it only results in overfitting of a system.
I don't agree. My stop is actived in less than 5% of all my trades. So to me it is a "last saving belt" if my system fails or if i make a mistake. The most important thing is to create a system that works without stops and without optimizing after backtesting. Your system should perform well as it is and under all circumstances. Only after that you have to think about stoplosses.
Stoplosses break scenarios that could have resulted in a positive result, and in consequence reduce your portfolio performance. They only work when the distribution of your returns approach something like a bi-modal distribution, where scenarios present a strong persistence. What you've described in your post is a typical overfitting. Sorry.
My testing with +1000 trades proved the contrary. My return improved. If i'm stopped out of a position that finally would have become profitable i will be in again before any damage to my return is done. If i'm stopped out that doesn't mean that i cannot take a new position in the same direction after i was stopped out. I just have to look for a new entry point. Stoplosses break also scenario's that could have resulted in heavy losses, and in consequence improve your portfolio performance. If i understand your logic you say that if you are stopped out that trade is gone and the profit is lost. The discussion is too theoretical. You have to see first what the system does before you can conclude anything. Some systems give much more, or more reliable signals, so there the influence of stoplosses will be different from lousy systems. As i'm not that smart and can barely read, write or calculate i had to use an approach that was adapted to my intelligence. So i started with a logical but practical approach. At my first year at the university i had a prof in math that could calculate things we never heard of before. When he was asked if he could predict the stockmarket he told us that it was impossible. As mathematician he is 1000 times smarter as i am, but he's not able to make 10% of the profits i make.
Your prof wasn't really smart as he forgot to tell you that maths haven't anything to do with a single case, and especially yours. IF IT WORKS IN YOUR CASE, THEN THAT'S GOOD FOR YOU, BUT : 1) THIS DOESN'T PROVE WHAT YOU'VE DECIDED OR IMPLEMENTED FOR YOUR TRADING WAS THE BEST WAY TO DO IT 2) THIS DOESN'T PROVE THAT IT WORKS IN GENERAL SO PLEASE STOP COMING HERE AND TELLING EVERYONE THAT THINGS ARE LIKE THIS OR LIKE THAT BECAUSE THEY ARE (OR YOU THINK THEY ARE) FOR YOU. In conclusion, if it works for you, then you're happy but don't claim that things are like this in general. You are in the very seldom cases (perhaps, as it is impossible to tell without knowing your trading method, which, I have to tell you, I don't really care about) I was refering in my previous post.
Well the fight is like between two guys I know: - one sticks to his long term proven strategy, listen to his technical analysis, never goes out of it, keeps the line of approved methods - and makes his money from the market - second has some basic streategy, bus on his words "he feels the market" and changes his buys and sells often based on intuition. And his revenue (as a percentage frombase) is higher Well as I am a poor newbie, I like to talk to them and realised, that NO method can work generally, you just have to find you the one, that you are happy with.