The general idea is CM does not appear to be working as expected for WW. What I'd like to do is get some data testing the CM proc rate for various spammable spells, since those are the easiest to check. The list of spells and general methodology is copied below, but see the link above for more details.
List of Spells:
For those willing to help test, pick a spell and rune and record DS casts and fight duration, as well as attack speed and crit chance. Your dps doesn't matter and 1-2 fights should be fine for the test, so don't worry about carrying out 20 tests to average over. Preferably do the fights against Ghom, and run to a corner at the start of the fight, wait for him to reach you, then spam the spell of choice while refreshing DS as often as possible.
Some initial results are in courtesy of some players posting on the official forums. Sig spells seem to not have any unexpected behavior with CM. The effective coefficient of WW seems to be impacted by not just IAS but also CC. The best theory brought up so far is a cluster theory, where the game takes all your attacks over a 0.5s period and adds them together to display the damage done. Crits are bottled up into that total but the cluster theory is that if any of those attacks procs CM, they all do. The theory is gaining promise by testing that shows CDs changing somewhat rapidly, as opposed to more smooth changes that might be expected over fights.
More testing of WW for various APS and CC values is going to be done and I'll post here when I get more concrete results. See the link in the first post for more discussion.
By all that is holy, this could potentially be a breakthrough discovery if the right evidence surfaces. Unfortunately, I'm very right-brained and all the mathematical skills that should be present in the left side of my brain are virtually nonexistent.
However, being as right-brained as I am, I will offer my creative two cents whenever the opportunity arises. Will be looking forward to the developing news here.
apo on the official forums did some very low APS testing of WW and RoF. His findings are extremely strange. In short, with low APS he's seeing more CM procs than expected LoH TICS. WW is showing similar behavior in that at 0.65 APS he's seeing about 10x the CM procs as expected.
TekkZero did some WW testing and found a noticable difference in what is best described as effective proc coefficient with different crit chance and APS. At 2.59 APS and 67.5% crit, he got an effective coefficient of 0.2016 but at 26.5% crit the coefficient was 0.2586. To compare with apo's test, at 43% crit and 0.65 APS his coefficient is something like 1.25. Around 1.75 APS it was more like 0.4-0.6. That is compared to the known LoH coefficient of 0.125.
So in short, something really strange is going on when it comes to WW and RoF CM procs or just crits in general. Unfortunatelly we have no idea what exactly is going on. The previous idea of clumping doesn't hold for lower APS results and no one has any ideas about just what might be happening.
What we want to do is run a bunch of tests for various APS breakpoints and CC levels to get some sort of correlation. We might not be able to model it completely but with sufficient data we can look into some empirical fit that will help us better understand the interaction. The main issue right now is time. Specifically there's only a few of us looking into the matter but data collection takes time, and doing multiple runs, plus recording and reviewing takes more time. If anyone wants to try helping, we would love some more volunteers. I personally have almost no spare time at all so data collection isn't something I can do for a while.