Predictability can indeed compromise the confidentiality aspect of the techniques being used for the show. It is not only intuitive, it is even supported by the probability theory. I know that the show is being coordinated by very competent people, but if I might be so bold as to suggest an improvement, I would like to say that introducing an element of randomness can reduce predictability significantly. We might, in fact, go further and associate randomness not just with individual events, but also with schedules. And the schedules themselves can be varied in timing, duration as well as length on the same principles. In other words, we can mix purposeful individual events (which have to occur at some specific time, either decided in advance or based on observation) with schedules that are of varying length, varying duration and continue for a varying number of days. These randomized schedules can serve as a camouflage for the purposeful events which require some confidentiality. As an additional benefit, their randomness can itself become purposeful in the same way in which the events are purposeful.
The problem can be formally studied in terms of the probability theory alone, but while implementing it in practical situations, we also need to take into account pragmatic, psychological and behavioral aspects. The latter are harder to study and require prolonged exploration that might require considerable resources. Ordinarily, i.e., under conditions different from those of the current decade, it would have been difficult to get support for such studies for a show of this kind. Fortunately, at present, we are in the midst of a security goldrush, otherwise known the terror goldrush*, that will most probably last for a long time. Therefore, it should not be difficult to get the required support. But wait, does not the said show owe its existence to the same goldrush? I think it does, but even so, we should try and get the required support and utilize it efficiently for the studies mentioned above.
* It really has terrorized the local-Truman. Ha! Ha!
So, my humble suggestion is that we should conduct these studies as expediently as possible and reduce the predictability to which the existing techniques are susceptible. Confidentiality is an important aspect and we should do everything we can to avoid it from being compromised.