key, there are more than one metric to use and the PE has it's value as a measure but there are other psychological, mathematical and behavioral measures as well and are completely different from PE in determining under/over value.
My measures says the S&P is over stretched and that there will quite a correction. Perhaps I'll tip the pivotal point for you when I believe it's reached and see whose system works best. That wasn't the nature of my post though. Say you were to determine over/undervalue by means not associated by direct PE but by, say, aggregate profitability of an index sector using an inverse measure from a parallel metric.
It becomes rather mathematical from that point. An Inverse etf/n is not necessarily the reverse of the etf/n for example. The whole point of the exercise is to try an pinpoint price fluctuation and in your case PE is just too broad to be reliable.
