That one's good for a chuckle. I can relate to the supposed infallibility of "the meter" and to the ease with which time, and battery life, slips by. I've gotten in the habit of dating batteries with a Sharpie marker when installing them to try to dispell the sense that I'm constantly replacing them. The facts prove otherwise in most instances (excludes my daughter's Walkman). When the battery in my Fluke meter expired recently, I would have sworn I just put a new one in it within a year. It was actually a little over nine years. That was almost exactly the lifetime of the only other 9v battery I've installed in it in the 18 years I've owned it. It has a "low batt" indicator, which I've never seen. I guess I've been lucky just to find a blank readout when the battery was gone.
I work in a highly regulated industry where I often scoff at some of the things we have to do to support the validity of data we generate in testing. Your experience is a good example of why we do need to be rigorous.