I looked at Roger's output file for all the M5s following M8, sorted by angular distance. First, some of the M8s have not M5s following within 2 deg-this seems unlikely, so must be an incomplete catalogue for M5s. I looked at 3 of the largest peaks 78 deg, 102-103 deg, and 155 deg. Only 102-103 corresponds to the FFA distances. The great majority of the number of M5s at a distance are located within 100-200 km of each other-meaning they are aftershocks of another large quake or swarms. Without wading into the main catalogue, I can't tell if they are aftershocks if a main quake occurs before the 14 day window.

So, for 78 deg: there are 5 locations quakes are triggered after the 27 main quakes, but a total of 31 M5+s at this angular distance
(26 in one swarm)

For 102 deg: 6 locations, 10 quakes total

103 deg, 10 locations, 29 quakes total
(20 in one swarm)

155deg: 9 locations, 41 quakes total.
(29 in one swarm)

So, to see the significance of following quakes, we would have to compare number of locations triggered with M5 or larger, not number of quakes. And quake within a few deg of another would count as 1 location (if the M9.5 1960 Chile earthquake had followed a M8 by 14 days, then just the aftershocks of that would be spread out over 1000 km and show a huge broad peak at a range of angular distances........)

So, are FFAs discredited? Not yet-there may be something there. But, so far it does not seem like there is anything major at the prescribed angular distances, except 102-103. And, I can't tell if 102 or 103 is within the range of chance...
Chris