Guys, the issue here is that some recent production 9s shoot well, others shoot less well/really bad.

Now, some of this could be explained as tolerance stack. For every dimension, there is a nominal dimension and a tolerance +/- of allowable variance. If you get a slide with breech area tolerances on the maximum limit and a barrel on the mimimum limit, accuracy is going to suffer.

Another possibility is that there could be one machine doing barrels that are out of tolerance or (if S&W isn't rifling their own barrel blanks) a supplier that is 'having issues'. The gent that reports a barrel throwing rounds 5" out to one side suggests that particular barrel has a bore that doesn't travel down the center of the barrel. This is something that very occasionally happens (although I'd really like to know how!) in a wide variety of handgun brands. Years later I recall a Colt 38 Super barrel whose chamber was 0.040 off center to the left.

The unlocking early claim was discussed in another thread. Any movement of the barrel breech is not unlocking. Only downward movement in a vertical plane is an unlocking action and that requires about 3/16" movement of the slide. Or at least it did in B-F serial series pistols.

Looking at that article on the 10-8 site, the original rifling spec has produced excellent accuracy in other model S&W pistols, that indicates that rifling twist has minimal significance in difference. However, it would appear that S&W bean counters decided to use one machine program to do all barrel profiles after development of the compact, regardless of end use. That would seem to be an error. The chamber area profiles of the original design are very similar to those used in the 1911, the P35 Browning and other pistols like the earlier S&W designs. These didn't have significant accuracy issues (or any issues-the nattering about stress risers to the contrary). It would seem that the easiest answer would be to return to the orginial barrel profile for full length and plus length pistols. There is another possible mechanical issue though: is there also a difference in locking blocks? Kinda doubtful, but slightly possible.

BTW, I can't speak for S&W, but one of the things we (plant maintenance) did during shutdown when I worked in a factory was a whole lot of machine tool rebuilding and replacement. Machines identified by QC as producing out of spec parts were taken off line when discovered and brought back into spec before being placed back into production. It would also be a good time to implement a decision to resume production of the old/another barrel profile for other than compact if they feel there is a market driven need that would justify the cost.