[quote=Return" or the $ figure ... Patrick :mrgreen:

You can't get an absolute answer, at some point you have to use statistics :D Always make sure you have enough data to satisfy your basic requirements but you have to make concessions at some point. Patrick :mrgreen:

6 sigma ... are you trying to be more cryptic than me :D You know how it works now, so you will have to decide how you want to use the system tester based on that info ...

:? I have been searching for this very same topic because I would really like Metastock to be able to calculate the testing results even more realistically. If we were able to say in the date range use data from date a to date b and then also just have another testing parameter that says start testing from date c to date b, then we would have the answer. Can we request this for a future version of metastock? Coding wise this can surely not be a big task. Out of market timing for example in the results section is totally messed up. I test over 550 days, I use a 200 MA, so really I am testing over 350 days. The lingest out of market timing comes out as 200+ (no surprise) and then surely that huge hit makes the average out of market timing useless. Buy/hold comparison and all those types of information looses meaning to a degree does it not? Are equis making any enhancements to metastock and if so when is the next release. Do you have access to people who decide what goes into the new releases?