I have been working on similar lines. Besides the WS statistic, I look at the value of Spearman's rank correlation coefficient (SRCC) for the data within a night. This measures basically whether magnitudes are more or less randomly spread over a night, or whether e.g. bright and faint mags tend to lie together. The latter is an indication of regular variation. Short period variables have a large WS statistic and a highly significant SRCC for all (or most) nights in V and I. Some new short period variable stars in the July data set, found this way (I have not checked all possible variables at this moment, I haven't paid much attention to the WS < 4 range yet):
At lower amplitudes (and lower WS), there is HIP 83986 = HD 155229 = GSC00407-00415 (WS=4), with an ascending and descending branch.
In this way I also found the known variable V2377 Oph (WS=7) = HIP 85944 = HD 159356 = GSC00992-01402, 2 nights: an ascending and a descending branch.
In contrast, for the Mira star SS Her (WS=70), the significance of SRCC for the I mags is low on the two nights, indicating long term variability.