My opinion on how to determine reliability has changed over the years of shooting. At first, I did it by successful rounds cycled and fired, mostly FMJ with small sample tests of JHPs. Then I got into IDPA and saw things differently.
Firing at a static range that most shooters have publicly available vs. an IDPA stage are very different. Most indoor ranges have rules on rate of fire, and I agree that it's an acceptable policy without knowing everyone's firearms experience. A "troubled" pistol may work fine during slowfire. But at a rapid pace, perhaps the magazine springs aren't keeping up. I've seen guns that work fine on a bench absolutely choke in an action shooting sport, where rate of fire can be pretty high.
So nowadays, my criteria is 100% flawless function throughout a full match or two. I highly recommend that those who carry use their carry weapon in IDPA. It's a good way to check for bugs or find a part failure. This is also where I see a lot of "limp wrist" failures in glocks, and is why I will not use something so temperamental. This "trial by fire" is how I determine the trust in a pistol, and no longer do I worry about firing a certain type of round. Check that it feeds, but don't shoot 200 rounds of it.