I did. Even streamed it with a real time view of an oscilloscope and thermocouple on the coil. The coil gets warm for sure and the resistance in the coil goes up when it gets warm but just a couple ohms. I looked at the pulse to the coil and it was always about 37 ms independent of the coil temperature. The voltage across the coil also didn't change over time. At the time I was doing it I figured I would see the time the coil was energized getting longer as the coil took longer to get to the limit switch but this never happened. Thus, I was never actually able to quantify any weakness over time. In hindsight maybe limit switch isn't used for changing the pulse duration like it is on some older games.
Like others I have felt on occasion that the coil got weaker with time. I felt this was especially the case with the TK lock on NST. I have not noticed the effect on other new Sterns but I know others believe it is there. I don't have an issue with my NST anymore after doing a number of fixes and I actually had to turn the powder down it get it to work properly.
I think the key would be trigger off the coil then use a second channel of the oscilloscope to monitor an accelerometer on the flipper or the end of range switch to see if the performance changes over time. Maybe I'll do this at some point, I'm just not as motivated at the moment because I feel like I'm not experiencing any issue.