there are significant systematic errors in RTOW charts
Ought not to be any significant errors at all .. certainly one applies some conservatism while running up the RTOW data but it is only a very minor consideration.
The only situation I could see where a problem might arise is if someone ran up a quick and dirty RTOW data set .. but, in that case, why not let the crews use the usual general takeoff data protocols as there would only be justification to incorporate the errors to get you out of a jam on the odd occasion ?
Might one enquire whence come your significant systematic errors ?