The reason that this quake is not a match is it is outside the distance boundaries
your program imposes. Allow your distance boundaries to be further for larger earthquakes
and you will find it. It does make some physical sense to allow longer distances for
larger quakes, as they may rupture up to several hundred km of fault segment - far
more than the allowed 3 times the 35 km range limit. No sure how this could (or
should) be done, though. Still, an event of Mb 5 at 110 km from an epicenter on the
date in question is probably a more meaningful fulfillment of a forecast than an
event of 2.5 at 35 km from the forecast epicenter. You're right - the program not
checking this event against Petra or Don's forecast doesn't seem quite right. I am
certain it would have a lower probability than the ones currently listed for Petra.