The Post-Upgrade P320 Question Nobody Really Answers

I'll bring what I know from the department side. We went through the voluntary upgrade in 2018—issued P320s across patrol and detention. Our armorer documented every FCU swap. Post-upgrade, we had *zero* out-of-holster discharges or unintended fires on the upgraded guns through 2023. That's roughly 280 duty weapons with full contact. Holster draws, retention drills, low-light scenarios, the whole rotation.

The gap you're pointing at is real, but I'd separate two things: institutional silence *about liability* is different from institutional silence *about the fix working*. Sig won't say "we had X failures per thousand" because that's discovery material. But they *will* document the change itself—metallurgy, geometry, testing thresholds. That's actually available if you dig through the FCU service bulletins.

Our qualification data shows no pattern change between upgraded and older stock either—same failure modes (shooter error, mostly). If the original FCU had a systematic design flaw showing up in duty use, you'd expect to see it in a large enough sample. We didn't.

That said, I agree with your practical advice. If I were buying used duty-adjacent, I'd verify the upgrade and price accordingly. The institutional silence on *scope* is fair criticism. But from the armory end, the fix appears solid. Not the same as transparent—you're right on that.

bluegrass.leo's department data is solid—that's a real sample. But we need to separate what it actually shows from what we're inferring it shows.

Zero out-of-holster discharges across 280 upgraded guns over five years is meaningful operational data. It tells you the revised FCU geometry functions reliably in duty conditions. That's the documented part, and it's worth weight.

What it doesn't tell you—what *no* field incident count can tell you—is the actual failure rate of the original FCU in the broader installed base. bluegrass.leo is right that Sig won't publish X failures per thousand because that's discovery-adjacent. But the *absence* of documented incidents post-upgrade is not the same as proof the original design flaw was rare or isolated.

Here's the distinction that matters legally: Sig settled the NYPD incident and ran a *voluntary* upgrade program. Voluntary programs exist because the manufacturer identified a risk they deemed addressable without a mandatory recall—which usually means either the failure mode was limited in scope, or the liability calculus made a recall more expensive than settlement plus voluntary uptake. We don't know which.

The OP's central point stands: institutional silence on *scope* creates a gap between "the fix works" (supported by bluegrass.leo's data) and "the original problem was understood and disclosed" (not clearly supported by public record). One is engineering fact; the other is disclosure fact. They're different categories.

For someone buying used, that gap means: verify the upgrade happened, negotiate accordingly, and move forward. That's practical. But pretending the gap doesn't exist is pretending we have transparency we actually don't have.

Not legal advice, but disclosure patterns matter when you're assessing institutional credibility on safety claims.

Let me break this apart, because this thread is doing what internet arguments do—stacking different kinds of evidence and calling them the same thing.

**What are we actually asking?**

bluegrass.leo has 280 upgraded guns, zero incidents. That's operational data. ctpistol correctly points out that zero incidents doesn't tell you the original failure *rate*. Both are right, and they're answering different questions. The problem is this thread is treating "Sig won't disclose" and "we see no failures" as if they prove the same thing. They don't.

**But here's where the burden of proof actually sits:**

If you're a manufacturer and you issue a *voluntary* upgrade program after a legal settlement, you're already admitting something needed fixing. The question isn't "was there a problem?"—that's settled fact. The question is "how *widespread* was it?"

And that's where I'd push back on the framing. The burden of proof isn't on Sig to prove the original design was flawed—they already proved that by fixing it. The burden is on *you* as a buyer to decide whether you want a gun with a documented design revision and an unknown incident history, or something else.

**What actually matters for your decision:**

If you own an original P320: upgrade it. Takes an afternoon.

If you're buying used: verify the FCU is upgraded. Ask the seller for documentation. If they can't produce it, walk or price it like an unupgraded gun.

If you're choosing between platforms: the P320 works fine post-upgrade, but if institutional transparency bothers you, buy what gives you peace of mind. Training matters more than this anyway.

Don't get stuck in the "how many really failed" argument. You already know enough to act.

bluegrass.leo's data is good—280 upgraded guns, zero incidents. I'm not going to pretend that's meaningless. But let's be clear about what we're actually measuring here.

Zero incidents post-upgrade tells you the revised geometry works. That's real. But the framing in this thread keeps treating "no documented failures" as evidence the original problem was isolated or minor. It's not. It's evidence of data suppression, not incident rarity.

Sig settled NYPD. The Army wrote complaints. But here's what doesn't get talked about: the P320 drop-fire incidents that *didn't* hit the courts or major departments. The range reports. The one-offs that got returned for warranty service and documented as "customer error" or "training issue." That data exists inside Sig's systems. We don't see it because manufacturers aren't required to publish incident patterns from voluntary upgrade programs.

I can cite the 2015 reports. I can cite the 2017 out-of-battery incidents from concealed carry users. I can cite the Crane incident documentation the Army filed. None of those "prove" a widespread epidemic—you're right on that. But they *prove* the problem existed at multiple levels, not just NYPD, not just institutional use.

ctpistol nailed it: voluntary program means Sig calculated settlement plus uptake was cheaper than mandatory recall. That's not a vote of confidence in design transparency. That's a financial decision.

So when gulfcoast_ops says "you already know enough to act"—yeah, you do. Upgrade the FCU. But don't pretend the upgrade proves the original design was a minor issue. It proves Sig identified a flaw, fixed it quietly, and moved on. Those aren't the same thing.

Honest: everyone in this thread is circling the same fact and missing the mechanical distinction that actually matters.

The upgraded FCU isn't just a geometry tweak. It's a different trigger bar engagement profile—the contact surface between the firing pin block and the trigger bar changed shape and angle. That's not a tuning. That's a redesign.

So here's what I think is getting lost: bluegrass.leo's 280 upgraded guns with zero incidents doesn't tell you whether the *original* FCU had a widespread failure pattern because those guns *aren't running the original FCU anymore*. ctpistol and p320.hater are right that absence of data isn't proof of rarity. But gulfcoast_ops is also right that you don't need the absence to disprove anything—you need the *presence* of the new design to work, and it apparently does.

The slip in the conversation: everyone's debating whether the original problem was big or small, but the actual question that matters for carry is whether the *current* gun is reliable. And on that, bluegrass.leo's duty data is directly relevant. Five years, 280 guns, duty draws and retention work, zero incidents. That's the sample that applies to you as a carrier.

Where I'd push back on p320.hater: the framing that this proves "data suppression" asks us to infer hidden incident counts. I'd ask—do you have evidence of specific incidents that didn't surface, or are we inferring from the structure of the settlement itself? Because those are different claims, and I'm curious which one you're actually arguing.

Right for: upgrading the FCU if you own an original. Wrong for: pretending the debate about original incident scope has direct bearing on carry reliability right now. It doesn't.