Jack Norris, a genuine "rocket scientist" from the Apollo space program, Technical Director for the Rutan Voyager flight, and author of the book Propellers Explained (and owner of the same Luscombe for 50 years), made a presentation to our local EAA chapter about tailwheel aircraft physics and flying techniques.
The key portion of this was an explanation that because of the human reaction delay and time lag between yaw axis acceleration, aircraft mass and its inertia, human reaction time, the delay between control input and result, control deadband near neutral, and probably 10 other things.... by the time the pilot is able to counteract a swerve or side gust, the timing is all f***ed up and it's far too late to be delicate.
If your Luscombe is starting to swerve left, and you gently put in small and increasing amounts of right rudder, hoping to not over-control it, so you keep increasing the right rudder control input until the airplane starts coming back toward the center. But by that time the airplane is actually already starting to come back through the center and into a right swerve, and then by the time you realize that and so you start gently putting in more and more left rudder. But by the time you see THAT correction take hold, the airplane is already swerving back to the left, and so on.
Essentially, a "gentle" pilot is always a half-cycle behind the airplane.
According to Jack, the only way to avoid this trap is to understand how much the time lag screws it all up, and apply a more significant corrective force much earlier. It's easy for us to think that to avoid a groundloop you should be gentle and only use enough rudder to keep the airplane straight. However, this gentleness is what leads to the swerving back and forth, with the pilot's timing a second behind where it ought to be.
So Jack coined a term for the correct technique to not have this problem: "Punch it and let off". When the airplane starts to swerve left, an immediate application of right rudder, and plenty of it, but then immediately go back to neutral. Because of all the time lag and inertia, you're already "behind" the airplane, so put in everything it needs as soon as you can, and then let off the correction before it can try to contribute any energy toward a swerve in the other direction.
This is NOT intended to encourage pilots to over-control or take liberties with the airplane, or be "ham-fisted" on the controls. My point is that a highly educated and experienced guy figured out a useful technique to address a common problem people have in ground-unstable airplanes.
I would be interested to hear if any of the high-time experienced instructors ont his forum have come to similar conclusions as my friend Jack Norris did.
.