1) Directional propagation of signals.
It's a possibility, but a bit like discussing how many legs a Woozle has. Either way, it seems most likely that main shock and aftershocks should share the same properties, i.e. if one is undetectable at a given location so are the others.

2) Signal to magnitude
Sure, you can argue that a precursory signal has no scaling with magnitude, but in that case you have no way of telling whether the impending event is a 3.5 or a 7.5 and the system is useless. Dennis claims he can predict magnitude, which implies some sort of scaling. It need not be linear, but the Hector quake is still 100 times larger than anything Dennis has handled before.

Roger