Well, just for the sake of understanding, as I am not the one who read any spectral analysis on oil samples, but have seen the systems and looked at a few graphs/analyses.
I agree with most of your statement, and since we are at the very most doing oil samples every oil change (or maybe more if you're nuts), I also agree that the frequency of monitoring would be too low.
However, if one were consistent, given engine break in, I would be curious to see what samples looked like time after time (hmm, light bulb), and also with other cars would love to see if there were, in fact, a correlation. As I stated before, we would draw samples every 10hrs on jet engines, SPECIFICALLY to look at metal contents in concern for bearing condition. There are obvious differences in the lubrication systems, mind you, so I know that there are variables.
All I am really saying is that if you draw the sample properly (less concern of particle stream, consistently, and manage other variables to the best of your reasonable efforts, "useless" is just a bit too absolute. . . for my taste (also my opinions). Perhaps I am completely left field, but there is definitely something to be gleaned. I know it isn't specific, but the fact that it is sensitive is maybe the downfall you're getting at.
I suppose I will formally start looking into oil analysis to gain an even better understanding.
Supposing you're correct that it is useless in that regard, this is certainly the way Murphy designed it :M5thumbs: Surely it couldn't be a way to avert disaster :flag: