In case you can't watch the video, this guy was able to reproduce an accidental discharge several times by putting a tiny amount of input on the trigger and wiggling the slide. He used a screw wedged into the trigger to consistently put the trigger up to the wall so it's hard to argue human error anymore. His findings seem to indicate that if you have input on the trigger, even if it's not enough to fire, and the slide is bumped, the gun can fire (at least for his particular gun).
Hard to say how common this will wind up being since there are tens of millions of P320/M17/M18s out there. Still...more bad news for Sig, and even a one in a million chance is enough for most folks to not risk it. I have one and, while I already never considered carrying it thanks to the lack of an external safety, I'd never even risk holstering it for matches or practice now.
The evidence really seems to be pointing to a low tolerance for out-of-spec parts in the P320 design. Any gun manufacturer is going to be incentivized to cut costs over time, so that's a really bad combo.
It isn't at rest, the screw basically holds the trigger at the wall. I think he says that near the beginning of the video. All the take-up is out, so this is simulating what happens if your trigger has some pressure on it, but not enough to push it past the break.
Some pressure can be put on the trigger easily in some holsters, which is why I never felt comfortable carrying the thing. Even if I knew for sure my 320 didn't have this problem, it was still too much of a risk to me that if something snagged the trigger there'd be no external safety to make sure it couldn't discharge.
The M17/M18 does have such a safety, but my understanding is that the safety is just on the trigger itself. If this problem is real, it's possible that the same thing could happen anyway; the striker is always in tension and if the safety was a bit out of spec then the trigger could be put into the same position as shown in the video.
The video is not terribly scientific, but hopefully some of the other gun YouTubers will try to corroborate it as well and more data will come out. MoistCr1TiKaL made a video about it a few hours ago, so this is officially a mainstream issue that gets clicks. Just from the collections on the dozen or so gun channels I'm aware of, there's a sample size of many dozens of 320s to test with.
Around 8:40 he makes a point to take measurements at rest and at the wall before including measurements that are somewhere between the two with the screw in place. I recall there being mention of the trigger only being “barely pressed”, not at the wall.
It’s definitely not very scientific, he is measuring the distance between two contoured objects with the firearm “floating” up in the air, so there’s a ton of potential for variance in those measurements.
I would have preferred to see the firearm laying flat during measurement so that we know both the tines of the caliper are against a flat surface.
I think a big problem is that Sig has been so obtuse about the problem that basically a random guy doing garage testing has become the only public and layman friendly source of anything even approaching transparent attempts to diagnose the issue and test the diagnosis.
100%. Sig's handling of this is, without a doubt, going to be a landmark case study in how not to handle public relations. People getting communications degrees 50 years from now are going to read about it.