Speed is relative to the site, the coil and the interpretation of the user. The term slow, to me, might mean fast to someone else. For example, my typical sweep speed in a park is slower than in a farm field. In parks I might take two seconds to go from left to right. And another two seconds to go back right to left. That is based on a small DD coil, 40 inch sweep and an overlap of at least 1/3. The coil choice, sweep speed and overlap is based on how I want to approach a site with multiple trash targets in the mix. In a field, where the ground level fluctuates more and targets are not as frequent, I speed up that sweep and try to overlap my swaths more. Especially if I am using the 9-inch concentric coil. The coil size and sweep speed allows me to cover more ground in a given period of time. And overlapping my swaths helps increase the odds that I am covering the areas as best I can.
I believe that sweeping the coil at a moderately quick speed does result in more finds. As your experiment illustrates, moving the coil too slowly can cause the detector to not react to the target. Afterall, it is a motion detector. And if your normal "hunt sweep speed" were as slow as the speed used in your test, you wouldn't have gotten that initial signal that made you stop and investigate, in the first place. I agree with Mark in that I use various sweep speeds to isolate and confirm a target. But as I said, if you're sweeping too slow in your hunt, you won't know if there is a target under the coil in the first place. My best advice to everyone is to do as you are doing. You can learn a lot about target response with airtests. But better yet, bury a few targets and determine the speed that provides the audio signal that best "catches your attention". Then, after the initial signal, vary your speed and determine what best isolates and identifies the target. As I said, sweep speed is relative to the site, the coil and the interpretation of the user. JMHO HH Randy