One conclusion that will result from precise observation of good dancers is that the interval between footfalls is actually a lot closer to being evenly distributed than the "SQQ" timing would seem to suggest. That's not to say it's precisely even, but that is the closest brief approximation. Looking in more detail, we find that the interval between footfalls corresponding to the last two quicks is the longest - typically a beat and a half if not more. The interval between the slow and the first quick is the shortest at just over a beat. And the slow itself is of medium duration, in effect whatever time is left between the drifted out preceding last quick and beat three on which we might reasonably assume the first quick lands.

(For representative English dancers the detailed implementation of "SQQ" is fairly consistent across all the basic figures, both feathers and turns. I could easily imagine other styles showing some variation by figure, but haven't taken the measurements to investigate that)

This at-first-approximation-even foot timing could easily support a relatively constant body speed, and to first approximation that is what we see.

However there is some variation of body speed. Generally we see a slight acceleration during descent and a clear slowing when gliding uphill over the top of the moderate rises. Opinions vary as to how much this should be - some will change body speed in proportion to the change of elevation as in the waltz (though of course the amount of change is drastically less given the gentler and more limited rise and fall), others will actively resist the natural acceleration on descent and so show even less change in body speed.