Rich, I'm with ya there. 
 I used the wizard, then edited the code to .02083 , then check with a hi power measuring optical comparator to get the actual mean pitch over about 8 threads, then adjust the .02083 slightly to get the best "average" results to account for the various variables (mechanics and sw). At one time, it was doing best at .0212 but only for about 100 parts, then it reverted back to very close to the starting point of .02083.

No, my goodness, never would pull out "instantly", that would be like an explosion ! Typical seems to be 45 > 90 by the attached printing.
Would not have to be 45 (wow, a quarter turn = Hi $$ machine).... or even 90 Deg. but would have to be no more than 360. 180 to 270 would be ideal.

Again in general.......threading will only be as good as the "LATHE SYSTEM" and the better you want it to be the more you need to pay since  you pay a lot for increasing accuracy.

I can swing the mechanical part of the "system", its the sw side of the system that is lacking. 

Thanks again Rich,
Russ