How Many Rounds Verify Handgun Reliability

[Hrfunk]

Here's one for discussion.

Howard

https://youtu.be/VUn_E2aB9Wg

[58marine]

Pi

[skin]

 As always, great video.

[czfwguy]

Nice video. Thanks for always taking the time to share great videos like this.

[xlazeruzx617]

Just checked this out, tyvm for information

[skin]

 Back when ammo was plenty, I would run 4 or 5 hundred to break an auto in and verify functionality. In today's world, might take some time.

[CZ_FANATIC]

I trust my Glocks straight out of their case. I've never had one malfunction in any way.

[CZMikey]

I trust my Glocks straight out of their case. I've never had one malfunction in any way.

I'm with you on this topic.  Having a bunch of Glocks and no malfunctions and watching my dept transition to Glocks for 1000+ officers with very, very few malfunctions.  I too trust Glocks outa da box.    
Well some ofc's could jam a baton and oddly they were the ones with problems with their Glocks.   
Now that doesn't stop me from polishing feed ramps and cleaning up slide and frame rails, but I have issues.   

I was raised on the 500 round break in period, so I stick with it, no matter what the gun is.  I do have more dollars than sense though.   

[Claymore504]

My SOP back when ammo was plentiful was I would run 500 rounds of FMJ and a 50 round box of the carry ammo I was going to use. I have not transitioned to a new handgun except for one recently and that is a P365XL. I have decided to start carrying it, but have only ran 200 FMJ rounds through it and 20 rounds of defensive ammo.

I do not completely trust any gun right out of the box. I do tend to trust some more than others. For example I would feel more comfortable running maybe 50 rounds through a Glock / M&P / CZ P07/P09 and then putting it into service, however with a 1911, I would need more rounds than that to have full trust.

With a revolver, I am good to go with running a few cylinders thorugh it. In the end, I would never take a gun out of the box new/used and put it right into service.

[MifflinKid]

Another well done and well reasoned video.

Thanks for making it.

[Hrfunk]

You are most welcome. Thanks for watching!

Howard

[Skookum]

The Non-Parametric Binomial Reliability Demonstration Test offers an objective answer to the question posed.  It is provided as Method 1 here:

https://reliabilityanalyticstoolkit.appspot.com/sample_size

You need to specify the following:
• how many failures allowed during your demo test,
• the reliability you require of your handgun system, and
• the confidence you desire in the test result.

Your handgun system is a combination of handgun + magazine + defensive ammo + shooter.  Change any element and you have a new system requiring its own test; extrapolate at your peril.

For example, testing a system by allowing no failures (to feed, to fire, or to eject), a reliability of 95%, and a test confidence of 95% requires 59 shots to be fired.  Note that with a semi-auto pistol the first shot from each magazine does not count, as the first round is not subject to a complete test trial (you feed the first round into the chamber manually).

Statistical confidence and reliability do not come cheaply.  For example, increasing reliability to 99% requires 299 rounds per system.  Even just 90% reliability requires 29 rounds per system.  When, in addition, relaxing confidence to 90%, 22 rounds per system are still needed.

[