Ok, I'm breeding a new Crobat and I've got a problem.
The parents: both with 31 all Atk/SpD/Spe - the mother was already successfully bred using the same method as described below (except with different date and delay).
The desired spread: 2009-07-09 16:01:14, 14 seconds, delay 628, frame 3, 31/A/31/23/B/A.
So I set the clock to 16:00 and soft-reset at 16:01, mashing A to load the game, B to cancel the journal and A again to receive the egg at frame 3 (like most eggs in the calibration phase), followed by the coin test. It did already produce the required thhhhthtth result twice, but the corresponding Zubats where different from the desired spread as well as from each other. Am I doing something wrong or is my luck just being horrible again? The fact that it still worked right with the mother, who had a 624/3 spread, really confuses me.
Thanks for help.