[old_tv_nut]

It looks to me like it's not really tightly focussed. To do damage, it has to really burn the phosphor, and it's not doing that.

Looking at the schematic:
1) I haven't spotted where point A 9 the B+ supply goes to, but it must be connected through some point to eventually feed C1.

2) The video signal is AC coupled to the CRT grid through C17; the grid DC is set by the cathode voltage of V13. I think this sets sync tips at ground (or close to that when the left half of V13 conducts.

3) The CRT cathode is connected to a variable DC by the brightness control.

When you turn the set off, I think the grid drops to zero (no more video), and the cathode voltage decays as the voltage on C1 decays. So eventually the CRT draws more current, but it's not enough to discharge the HV capacitor before the raster collapses. It seems to me it's designed that way, and that is that.

The focus voltage will also be collapsing, and my guess is it has gone to zero by the time the spot forms, so the spot is larger than during normal operation. I played your video at 1/4 speed to have a better look, and it seems the spot is large and not changing in size. Try that and see what you think.

Some sets were designed in such a way that the CRT bias voltage collapsed very rapidly, so the HV capacitance was discharged before the raster completely collapsed to a spot, but this apparently isn't one.