NOTE: This is not a question about coordinated flight in the pattern. There is no disagreement that hasty skids to turn the plane faster from base to final can and do kill. The question is about what happens once the plane is aligned on final and now continuous micro- and small-adjustments are necessary to precisely land the plane on the centerline.
While flying Condor the other day, I noticed that when I depend on automated turn coordination it is exceedingly difficult to align the (simulated) glider with the runway. The reason is readily apparent in that a typical glider has a bank rate of less than 30deg/sec, and thus bank corrections happen far too slowly to effectively steer the plane. The world is simply going by too fast for the slow rate to yield the precision required for landing. And from a theoretical controls perspective, this is a poorly damped oscillatory system, which are famous for being touchy to control.
And yet, when actually flying for real, a century of anemically maneuverable aircraft have demonstrated that high roll rate is not a requirement for safe landing. What then is going on?
I hypothesize that while on short final we use the rudder extensively for fine control, while keeping the wings appropriately level for the cross-wind conditions. In other words, we intentionally and unconsciously slip and slide the plane in small amounts, because coordinated turns in this phase of flight are slow and even dangerous.
However, this is merely an uninformed hypothesis, and there is certainly a large body of evidence which can definitely answer this one way or the other. In particular, this must be well understood for autolanding systems.
Can anyone point to studies which looked at the rudder movement on final approach vs. in standard flight?
P.S. Anecdotes are interesting, but I think these should be confined to the comment section. I think in order to really answer this question we need results from empirical studies.