Hi, just registered to contribute what little I can to this thread.
First of all, the OP is spot on with just about everything. I have some minor issues about enchants (mostly about Crusader being ignored) but more about that later; I'll tackle the misinformation and misconceptions in the numerous replies first.
A lot of people seem to have trouble understanding why 1% increase in crit is better than 1% increase in hit even if you just consider auto attack (provided you're not crit capped, which shouldn't be the case anyway). Let's have a simple thought experiment of 100 attacks done with a static 100 damage per attack and 200 damage per crit:
20% crit and 80% hit chance:
20*200 + 60*100 = 10000 damage
Adding 1% crit:
21*200 + 59*100 = 10100 damage
Adding 1% hit instead:
20*200 + 61*100 = 10100 damage
You can change the chance of hit and crit around and change base damage and it won't change the outcome, critting makes up for the lost hit. We are also ignoring some advantages that crit has such as Impale and not being subject to glancing blows, which would tip the scale clearly in favor of crit. And lastly remember, this is just for auto attacks. After 9%, hit chance does nothing for your specials while crit obviously does.
But isn't +hit better for rage generation than +crit? Well, you could argue that hitting more provides more steady rage income than critting more, but that's about as far as you get with that. When the rage formula is considered, 1 crit is pretty much equal to 2 hits (crit receives 2x the multiplier on the static part of the formula) and again, Impale and glancing blow concerns bring crit out in front in the comparison.
Bottom line: It's easy to see that 1 crit is better than (or at least equal to) 2 hits in most situations, but that's the trade you're doing when stacking hit instead of crit. So don't do it. Only thing hit does have going for it is that it takes less hit rating (~16) to gain 1% than it does crit rating (22) but it's not nearly enough when you factor in the instant attacks that don't benefit from hit at all (after 9%).
Another issue which someone definitely got wrong here is the actual nature of PPM.
PPM is a rather outdated way of describing a frequency of randomly activated enchants, since
a) instant attacks definitely alter the scenario
b) as do haste effects (and rating).
As a result, an enchant such as Crusader with a nominal PPM of 1 will in fact proc quite a bit more in any realistic situation.
WWS
In a fight that lasted under 8 minutes, I was healed by Holy Strength healing 22 times and gained Holy Strength 12 times; what this means is that Crusader procced 22 times in under 8 minutes (and 10 of those times the last proc wasn't out yet). That's not 1 PPM that Crusader is supposed to have; it is in fact closer to 3 PPM. Admittedly I don't always get a posterior PPM high as that but it's nearly always at least 2. Oh and yes, I have Crusader only on my Main Hand (Potency on OH); this wasn't dual enchants.
It should be rather obvious that the proc chance of a PPM-based enchant is calculated as a static figure based on your weapon's speed and NOT adjusted down when under haste effects (I've heard this might be "fixed" in a future patch though). But don't take my word for it, go test it yourself.
As far as enchants go, Crusader already came up in the discussion above about PPM but I'd still like to point out that it only needs to be 20 seconds per minute to outperform Potency, which it in most scenarios will, by a rather clear margin.
Also while I realize that Mongoose will outscale Crusader eventually, I'm not convinced that such point is easily attainable with current gear. I haven't done the math on it but the simple fact that rogues get 120 AP on top of what warriors get from Mongoose, it would be utterly out of whack for Mongoose to outperform Potency/Crusader without the AP component and I can't believe Blizzard would have fumbled the balance of the enchants THAT badly.
Lastly it's a bit sad to see how easily people swallow false theorycraft (25% block needed for crush immunity, haste effects lowering rage per swing; I could go on and on) without really giving the matter much thought (or testing) themselves. When it really becomes annoying is when they come out to defend their false dogmas against new thought or tests, STILL without actually having (or being able) analyzed the matter themselves.
Cheers, Rothion.
Edit: Sheesh I don't know what I was thinking, hopefully no-one already replied to the bogus numbers in my crit vs hit thought experiment. Should be fixed now
