Pistol optic mounting referenceOptic Mounting
Home / Guides / Red dot lost zero

Red dot lost zero? Find out why

A pistol optic that loses zero has almost always moved, been moved, or is being read differently — and the order you check those in matters. Answer one question at a time; each answer takes you to the next step, and every cause ends with how to put it right.

Question

What are you seeing?

Start with the mount, not the optic. It is the cheaper thing to rule out, and the one this site can actually help with.

Question

Did you witness-mark it when you installed it?

A witness mark is a line of paint pen across each screw head and onto the optic or plate. It shows movement too small to feel, and it splits everything below into two groups: the mount moved, or it didn’t.

One exception: GLOCK Gen6 ORSReplaces MOS on Gen6. Polymer spacer plates screw the optic directly into the slide, so the optic sits lower and there is no metal plate to work loose. Early field reports include optics rotating on the polymer without visible witness-mark movement — check zero often and re-torque on a schedule.

Start again

Question

Does anything move by hand?

Clear the pistol first: magazine out, slide locked back, chamber checked by sight and by feel. Then push the optic firmly side to side and front to back, and put a torque driver set to the spec on each screw — one that turns before the driver clicks was loose. Don’t go past the spec to find out.

Start again

Question

Which joint is moving?

A plated slide has two joints, each with its own screws and its own torque figure, and they fail for different reasons. A direct cut has only one.

Start again

Question

Did it move right after a battery change?

An optic that loads its battery from underneath has to come off the pistol to change it, so every battery change is a re-install. Of the optics tracked here, these 3 are listed as bottom-loading:

Each optic’s page lists its battery and where it loads, where the maker publishes it.

Start again

Question

Did anything else change since it was zeroed?

Different ammunition, a different distance, a different shooter, or a different way of holding it.

Start again

Question

Does it still miss from a rest?

Shoot a slow group from a rest or a bag at the distance you zeroed at, with the dot in the middle of the window every shot.

Start again

Likely cause

The optic’s own screws

  • Oil on the threads or the faces. Threadlocker doesn’t cure properly on an oily thread, and a film under the optic lets it creep. Degrease the screws, the holes and both faces.
  • No threadlocker, or a spent patch. Some screws arrive with a dry patch of threadlocker already on the threads, and it is generally meant for one installation. A screw that has been in and out needs cleaning and fresh blue threadlocker — never red — or replacing.
  • Torqued by feel. Each footprint has its own figure, below. “Snug” on a small screw in a thin slide is how it backs out.
  • A screw that’s too long. It bottoms out before the optic is clamped and feels tight while the optic isn’t — see that step.
  • Recoil posts not seated. An optic resting on top of a post instead of in its socket puts the recoil on the screws. See that step.
Optic screw torque, by footprint
FootprintScrewsTorque
Trijicon RMR#6-3215 in-lb
Trijicon RMRcc#6-3218 in-lb
Shield RMScM4 × 0.7 (some hosts #6-32)10–15 in-lb per plate maker
Holosun K#6-3215 in-lb
Leupold DeltaPoint ProM4 × 0.720 in-lb (confirm current Leupold spec)
Docter / NoblexM3 / #4-40 by model15 in-lb
Aimpoint ACROCross-bolt clampper Aimpoint manual
Holosun 509Tper Holosun kit15 in-lb plate / 20 in-lb optic
C-More RTS2 / STS2per C-More kitper C-More manual
Aimpoint Microper mountper mount maker
Trijicon RCRMounts from beneath the slideper Trijicon manual
Aimpoint COA / A-Cut2 rear screws + front tongueper Aimpoint manual
Holosun SCS (MOS)per SCS kitper Holosun manual
Holosun SCS (320)per SCS kitper Holosun manual
SIG Romeo1Proper SIG kitper SIG manual
SIG-LOCScrews from beneath the slideper SIG manual
SIG-LOC PROScrews from beneath the slideper SIG manual
SIG Romeo2per SIG kitper SIG manual

From the same data as the footprint pages. Where your optic's or plate's maker publishes its own figure, theirs wins.

Put it back on properly

A loose optic is a re-install, not a re-tighten. The same sequence as a first install, from the fit checker’s build sheet:

  1. Clear the pistol. Drop the magazine, lock the slide back, look and feel the chamber, then do it again.
  2. Remove the cover plate and set it aside with its screws.
  3. Degrease the slide cut and the optic base. This matters more than people think.
  4. Seat the optic on the slide's recoil lugs and check it sits flat with no rock.
  5. Fit the optic screws with blue threadlocker and torque to spec, alternating between them.
  6. Let the threadlocker cure before firing — check the bottle, usually a few hours to overnight.
  7. Confirm zero at your working distance, then re-check torque after the first 100–200 rounds.

On a plate, the plate goes on first, with its own screws and torque — the plate step has that sequence.

Start again

Likely cause

The plate, on the slide

  • The wrong screws. If the pistol came with a cover plate, its screws are in the same bag as the plate screws and they are not the same length.
  • Oil, no threadlocker, or torque by feel — the same three as the optic joint, with the slide’s own figure. 2 of the 34 plate systems tracked here publish a number (GLOCK MOS — standard & full size and GLOCK MOS — 10 mm / .45 frames: 13.1 in-lb (1.5 Nm)); the rest defer to the plate maker’s instructions. Your pistol’s page says which.
  • It doesn’t sit flat. A burr, debris or finish in the slide pocket leaves the plate rocking on a high spot, and the screws take the recoil.
  • A plate that keeps working loose however it’s fitted. Thin slides leave little thread to hold, and a steel plate holds up better than aluminium there. Fewer joints is better still: a pistol page’s “No plate needed” count is the optics that mount with no plate at all.

Put it back on properly

  1. Clear the pistol. Drop the magazine, lock the slide back, look and feel the chamber, then do it again.
  2. Remove the factory cover plate and set its screws aside — you will not reuse them for the optic.
  3. Degrease the slide pocket, both faces of the plate and the optic base. Oil under a plate is why plates move.
  4. Seat the plate on the slide, checking it sits flat with no rock, and fit its screws with blue threadlocker.
  5. Torque the plate to the slide to the plate spec, working the screws evenly rather than one at a time.
  6. Seat the optic on the plate's recoil posts, fit the optic screws with blue threadlocker, and torque to the optic spec.
  7. Let the threadlocker cure before firing — check the bottle, usually a few hours to overnight.
  8. Confirm zero at your working distance, then re-check torque after the first 100–200 rounds.

Start again

Likely cause

A mount in the rear-sight dovetail

The clamp or set screws that hold a mount in the dovetail are the weak point of that setup — check them more often than you would a plate. A mount refitted at even a slightly different angle to the bore moves the zero with it.

Slides tracked here with no optic cut, where a dovetail mount is the route:

Put it back on properly

  1. Clear the pistol. Drop the magazine, lock the slide back, look and feel the chamber, then do it again.
  2. Drift out the rear sight. Support the slide properly — this is where people mar a finish.
  3. Degrease the dovetail and the mount's foot.
  4. Fit the mount, check it is square to the bore, and tighten its clamp or screws with blue threadlocker.
  5. Seat the optic on the mount and torque to the optic spec.
  6. Let the threadlocker cure before firing — check the bottle, usually a few hours to overnight.
  7. Confirm zero at your working distance, then re-check torque after the first 100–200 rounds.

Start again

Likely cause

A screw that’s too long

Stop — don’t fire itA screw that is too long bottoms out in its hole before the optic is clamped down. It feels tight — a torque driver may even click — while the optic is still free to move, and on a slide the tip can reach the striker channel.

The usual way it happens: the screws in an optic’s box are cut to mount it straight onto a slide. Put a plate in the stack and the geometry changes, which is why the plate maker supplies the optic screws for a plated build. Screws from the wrong bag, or one set reused on a different stack, are the classic case.

Measure, don’t guess. This site will not print a screw length, because it depends on your exact stack. The fit checker’s build sheet has a screw-length check that works from two caliper readings of your own — the stack the screw passes through, and the usable depth of the tapped hole — and gives you the longest safe screw.

Start again

Likely cause

A screw that won’t tighten

Stop turning it. A screw that spins is either stripped or cross-threaded, and more turns make both worse.

  • The wrong thread. The tapped thread lives in the slide or the plate, not the optic, so screws from the optic’s box can be the wrong thread for your slide. Test one by hand before reaching for a driver — cross-threading an optic body is not repairable.
  • GLOCK Slimline MOS (43X / 48). GLOCK taps the Slimline MOS slide M4 × 0.7. A Holosun K optic will bolt on, but the #6-32 screws in the Holosun box will not thread into it — you need M4.
  • GLOCK Slimline MOS-K. Tapped M3, not the M4 of the earlier Slimline MOS. The two slim GLOCK cuts look alike and take different screws — this is the single most common cross-threading mistake on slim GLOCKs.
  • Docter / Noblex optics. Different makers on this one footprint ship M3 and #4-40 — check what is actually in your box.
  • A stripped hole. In a plate, replace the plate — plates are cheap. In the slide, it is gunsmith or manufacturer work. Don’t force in a bigger screw.

Every pistol page gives the slide’s thread where the maker taps it, and every build sheet names the thread for each joint.

Start again

Likely cause

The optic won’t sit flat

An optic should drop onto its recoil posts and sit flat before a screw is turned. Never use the screws to pull it down — that is the recoil going through the screws from the first shot.

  • A footprint that looks right and isn’t. Looks like a small RMR and is not one. Needs its own cut or plate; will not sit on an RMR or RMSc pattern. Trijicon RMRcc.
  • Posts the cut doesn’t have. A subset of RMSc: K optics drop onto RMSc cuts, but a K-only cut has no rear posts for an RMSc optic to index on. Screw thread differs from Shield's M4 — check what shipped in the box. Shield RMSc vs Holosun K sets out which goes on which.
  • Something in the way. A burr, a thread of debris, excess finish or a proud screw head in the pocket. Clean it and try again dry.
  • Pins or a plate not fully home. On a slide that locates the optic with pins or a plate, those have to be seated first.

Start again

Likely cause

The battery change was a re-install

Taking the optic off to change the battery undoes everything the install did: the threadlocker, the torque and the witness marks. Treat every battery change as a re-install — degrease, fresh blue threadlocker, torque to spec, new marks — and check zero before you trust it.

If you’re choosing an optic and this matters to you, a top- or side-loading battery avoids it; each optic’s page lists where its battery goes, where the maker says.

Start again

Likely cause

Something else changed

  • Ammunition. A different load can land somewhere else. Zero with what you carry.
  • Distance. A dot zeroed at one distance lands differently at another. Check zero at the distance you set it at before deciding it moved.
  • Shooter or grip. A group that moves with a new shooter, or a new grip, is usually the hold — see that step.

Start again

Likely cause

It’s the hold, not the hardware

If it groups on zero from a rest, the optic and the mount are doing their job. What’s left is how the pistol is being held and sighted: grip pressure, the trigger press, and where the dot sits in the window — keep it centred, since the window’s edge is where a small error in head position shows most. Work on it from a rest outward, and leave the adjusters alone.

Start again

Likely cause

The optic itself

If the marks haven’t moved, nothing turns by hand, the ammunition and distance are the same, and it still misses from a rest, what’s left is inside the optic. That isn’t something a mount can fix: contact the optic’s maker, and stop re-zeroing it until then. Say what you’ve already ruled out — it will be the first thing they ask.

Start again

For your pistol and optic

Pick your pistol and optic for the slide’s thread, both torque figures and a build sheet with the install order.

Something on this page wrong? Say what it should be. Corrections from someone who owns the pistol are why this is worth more than a spec sheet — and nothing here changes without a source, so bring one.