I have mentionned two things inside two implicit frameworks one that is stochastic, one that isn't. It seems that that you don't understand the term tolerance which in Statistical Quality Control means the minimum or intrinseque volatility of the system. The term tolerance is used in this domain because it puts more clearly the accent that you cannot violate it by asking more that it can do (it is not only for that but I won't enter into details). So the consequence is obvious (that's why I didn't say it): your stop must be function of this tolerance if you want to use a stochastic approach so <B>I'm saying the same thing than you in this case</B>. Now if you absolutly want tight stop it is not impossible but you must then use another kind of framework than a general stochastic one. For example as for my model it shows that there is some point that can be calculated 24h in advance that are <B>"never come back" zone</B> so the stop can be very tight (this is llustrated on my guide see http://perso.wanadoo.fr/harrytrader/_sgt/m1m2s3_1.htm the comment "This a zoom of the local minimum on DJI hourly (labeled MinDJITP1). Note that once the next level MinDJITP1[+1] is touched it becomes a "never come back" zone."). My model must not be the only one so use what you want but what I wanted to say is that it is not impossible to have tight stop if you use refined knowledge, if you want to content with stochastic knowledge then it is impossible due to the tolerance concept above and then you are right the stop must be function of the volatility. My model shows that the never come back zone are very rare so that if you try to pick them by random you won't have many chance to be successful that explains that it would be crazy to use too tight stop ("too" is always relative to something so here it is implicitly relative to tolerance or volatility) in a stochastic framework. In fact it is so many times that I have repeated the concept of tolerance for stop that I thought everybody has assimilated it so it's just one more repetition. Perhaps I wasn't enough explicit because in my head it was a continuation of posts I already made on that kind of subjects although they may be too old for people to remember.
I just saw you have only 5 posts so you must be a newbie and never read my former posts. Then next time before pretending that something is clueless ask before. As a former statistical control engineer I may surely not ignore what VOLATILITY is so thank you to remind me that basic term for the hypothetical case I didn't know what it is but be reassured that I'm not old enough to get the alzeihmer illness to have yet forgotten it
markets are stochastic period. i'm not sure if it is your thinking that is muddled or your use of english, or both. comments like "I have mentionned two things inside two implicit frameworks one that is stochastic, one that isn't." - muddled "It seems that that you don't understand the term tolerance which in Statistical Quality Control means the minimum or intrinseque volatility of the system." - muddled, huh? not my definition of tolerance. "The term tolerance is used in this domain because it puts more clearly the accent that you cannot violate it by asking more that it can do (it is not only for that but I won't enter into details)." - muddled "So the consequence is obvious (that's why I didn't say it): your stop must be function of this tolerance if you want to use a stochastic approach so <B>I'm saying the same thing than you in this case." - muddled, but thank goodness for that, we are on the same wavelength...i think? only with your statistical control engineering background you need to use 10 times more words to say it. after all that muddling, on a more positive note, i see you are lord of the rings fan. have you ever visited middle earth? i actually live there - www.newzealand.com
If your setup is right for high probability winning trade, then only wait for predetermined time (few minutes or seconds), and exit. Usually, good trade happens immediately. Also, size of the bet determines how much risk you want to take. Most of the time do not stay in the mkt, just observe. opportunity will fall on your lap, as you become more astute observer. For small traders like us, only few way to trade after recognizing the long term trend look for- breakout, or retracement, or s/r, or pivot/point, or some chart patterns, or volume analysis, and sometime news. So, try to be expert in one or two items and you will be alright.