I have been building on data 2013 and forward since that was supplied by default with SQ4. However on most instruments the result before 2013 is usually horrible either almost no trades or flat or loosing. Out of millions that passed IS and OOS including WFA only a handfull is profitable 2003-2013. So i broke out the 2003-2013 and put this on additional markets and with the exeption of a handful strategies of about 2 million generated they all look like attached. Another reason for me to think that something is wrong is beacuse if i add to addtional markets same period 7 instruments 2013-2018 my data banks quickly fills up with strategies that passed 2013-2018 on all additional instruments but they fail badly on 2003-2013.
I have not seen any big diffrence in strategies between 2007-2013 and 2013-2017 in SQ3 there was not a big change in the market behavior that i am aware of so thats why i think that the OOS function in SQ4 does not work.
Failure rate is the same building without OOS on additional markets when re-testing on 2003-2013 . If OOS works on SQ4 i suppose the learning is that a 2-3 year OOS has non what so ever impact on future performance neither have WFA or Montecarlo.
This also brings up that the most important feature is actually additional markets and the possebility to set another data period then the original build. Since it is no use testing on additional markets fro the same period of data since they are to correlated ( obviosly)
It seems most all of above is related to me only looking at the EQ charts. When I added colums and values for Additional markets I saw that it is the EQ charts that are not displaying correct data.