I’ve see some incandescent bulbs arc at EOL and form a gap, but I’ve only seen it with bulbs with C-9 filaments or another single coil design, never a standard A-19, reflector lamp, or G-25 with a CC-6 or CC-8 filament, I’ve only seen it happen on bulbs with single coil filaments.

I’ve never seen it happen on ANY double coil filament. Only single coil. Anyone know why that is? I haven’t used incandescent in a long time but I’ve seen incandescent lamps arc before with a bright filament spot, in fact, I once went to a restaurant and saw an arcing clear Satco 60W lamp above me, at the table, tho it was a long time ago it did have the C-9 shape.

Anytime I’ve seen an incandescent bulb arc, it used a single coil filament.

It also happens on coiled-coil filaments, in fact that is slightly more common than single coil due to the higher voltage gradient.

One possible explanation for your observation could be that for general service lamps, most of the bigger and more reputable manufacturers use coiled-coils wherever possible, and single coils are more prevalent among cheap manufacturers who do not care so much about quality.

Maybe this applies to lamps in different countries but I’ve never seen a U.S. coiled coil lamp arc.

I've seen here in Israel, coiled coil incandescent lamps that arced at EOL.

On 230V the arcing was way more common, I even do not think it is really avoidable with that high voltage, it was mlre of a matter who can design better fusing to prevent the bulb to disintegrate (in my childhood it was considered as rather normal for the lamps to explode at EOL in the larger appartment buildings where the mains had rather low shortcircuit impedance...).
But the original question was related to the 120V, where the arcing wasby far not that common...

Maybe with a single coil, sufficient part of the filament is far enough from the arc, so can act as a ballast and keep a current limited arc going

With a shorter coilded coil filament, the electrical field may tend to pull the arc to extend over the entire filament, so either blow a fuse, or melt a part of the filament that results in extending the gap between the ends, which then won't keep the arc going

On 230V the arcing was way more common, I even do not think it is really avoidable with that high voltage, it was mlre of a matter who can design better fusing to prevent the bulb to disintegrate (in my childhood it was considered as rather normal for the lamps to explode at EOL in the larger appartment buildings where the mains had rather low shortcircuit impedance...).
But the original question was related to the 120V, where the arcing wasby far not that common...

Ditto for 220-230V. Combined with a poor quality socketing glue used at Soviet time, that was fun! You flip the switch, a flash, crack, and entire bulb falls out to the floor heh). Circuit breaker tripping added some visual entertainment in the evening)

Removing the base remaining was some extra fun of pliers work ))