Quote from frostengine:
Maybe I am missing something here, but there seems to be an obvious mistake.
From what I gather you have your entrance at open +x... then stop at open. And you expect it to have the same % accuracy as what you figured when performing the intial calculations.. however that is impossible...
Simply because your calculations deal with something such as abs(open-high) and abs(open-low)... so your shortest distance and your longest distance variables are generated from above..
This does not give you any information to be able to say if price moves greater than open+x to enter and then to get stoped otu if it moves back to the open.. it should be if it moves past the open-LSD...
That is because its easy for the price to pass your entrance point, still go back below the open or further, not get all the way to open-LSD,then turn around and go back above yoru entrance point as much as your SDL...
do you get what I am saying? Am I describing this correctly? Bottom line, I dont get the stop being at open.. using OHLC bars you do not have enough data to know what came first... therefore yoru % of success for the trade numbers are way off... you should be using your stop for longs at open-LSD
Which this makes your risk for your trade MUCH greater, makes it more like 2*LSD instead of just one LSD..
comments?
read the thread through is said the built in stat loss is around 20% because it revert to open thus and stops from my data about 20% of the time thus making a 90% level 90%*80%= 72%
notice then later i said that while you learned that ...now we can fix a mistake in thought process and use tick data and run the stats for what you just mentioned..... thats the next step perfecting and removing 20% tendencies. while the system in the beginning does work it have a unremovable built in loss around 20% we take our idea a step further in the next few posts to really form the idea we were looking for .... this is an "approach" to helping create a working system that you know your odds. this just shows you how and gives you an idea to using stats.
now the info we gained is not in vain because we can now look at our past way of putting this in a spreadsheet and finding formation probs... showing us tendencies in formation its helpful.
then show you how to apply the same logic to tick data now you know your direct odds... this is a thread that builds on itself. IM not giving you a system to give you a system is to give you nothing. to give you an approach is helpful.