Originally Posted by Skarg (Post 2259135)

I think that indeed this:

Means that while the figure forcing retreat can choose which (more-distant) hex the retreated figure retreats to, it also qualifies the retreated figure to choose a new facing in that hex.

This is also consistent with the other rules that allow a figure to move a hex outside the movement phase (e.g. when retreating from a Push, or when rolling out from under a larger figure, or when disengaging from HTH).

It's also very easy, and often doesn't matter (but sometimes can, because it can determine who that figure engages, and who might be able to jump the retreated figure in HTH, during the upcoming movement phase).

Originally Posted by FireHorse (Post 2259123)

but what if they weren't facing the Attacker?

Originally Posted by larsdangly (Post 2259186)

the best argument I've seen in this thread is for a uniform rule that the player controlling the figure that moved picks the facing

Originally Posted by larsdangly (Post 2259248)

The flip side to this argument is that real historical fencing, which operated in an open space rather than a modern fencing strip, puts a lot of focus on working your way into positions that are effectively equivalent to attacking through a side facing in TFT. This would be possible if combatants who were forced to retreat were not allowed to change their facing as they do so. But, most clever thoughts about versimilitude make for terrible games. So, I don't advocate this sort of thing.