We have now had a month of posting 9-hole scores under the new system which is a decent sample size for my course as maybe 50% - 60% of the rounds posted by players in my group (25'ish players) are 9-hole scores rather than 18 and almost all of us play multiple times per week, every week.
The general consensus seems to be that indexes feel inflated. That makes sense considering the Expected Differential for the 9-holes not played is based on an average for players in the same handicap range. It eliminates the possibility of a 9-hole differential ever being posted as 2x9-holes of better than average golf which I am not convinced is the best way to approach it.
I am working under the assumption that the Expected Differential is an overall average for people in the same handicap range rather than an average of only the scores that counted for handicap purposes. It would be interesting to see if there would be any difference between an average 9-hole differential for the back 9 of 18 hole rounds when the score counted for the handicap calculation versus just the Expected Differential which is an average of all rounds(?).
At any rate, it seems like they are trying to make 9-hole rounds more comparable to 18-hole rounds but I am not sure they are approaching it the best way but of course I have no hard data to back that up.
Just off the top of my head I get the feeling that forcing players to post a minimum of the front 9 score and back 9 score for 18-hole rounds (versus allowing just the overall score) and then using each 9-hole score as a 9-hole differential would put things maybe more in line. ie. your index would be the best 16 of 40 x 9-hole differentials x 2. The overall 18-hole index then might need to be multiplied by a small multiplier, like 1.02 for an 'out of thin air' example, so that indexes remain more in line with how they have been in the past.