I do not want player stats simplified.

What we were finding is that in many cases things had just become so complicated that players were not making intelligent decisions based on their knowledge of how various stats benefited them. That sounds like what you wanted to have happened, but it really wasn't. Instead players would pull up a BiS list, or plug the item into a spreadsheet. If the answer was "upgrade" they equipped the item. We could pretty much just give the items a name and art and make all the stats not displayed.

If you were one of those players who picked up a piece and understood how armor pen vs. attack power benefited you or whether defense was still worth collecting after you had hit crit immunity, then you were in the minority. Hopefully we can get back to a system where eyeballing an item's stats has some chance of being right.

Q u o t e:
I'm sorry, but I fail to see your logic. If players are itemized poorly, it's because they don't care to do enough research to learn how to properly itemize or how to successfully itemize to fit their play-style.

I find many of these arguments to be elitist. This is not a "catering to the casuals" issue. This is an issue that even some of our most educated and math-savy players could not make gear decisions without the help of external tools. We're much less concerned about more casual players because the specific choice of upgrade will mean less for them (skill outweighs the contribution of stats for all but the absolute best players) and they often choose pieces based on things like art over stats.

I'm talking about very hardcore raiders who no longer could eyeball a piece of gear with any degree of authority. I might believe that some very few of you are so proficient with the math as to be able to eyeball the stats, but I think you are vastly overestimating the number of players who can or do. Remember, the formula to convert armor penetration into damage was so complicated that we had to take the unusual step of actually spelling out how it was calculated.

With this change will you still have to make hard choices about upgrading? Absolutely. Will there still be quasi-religious debates about which neck is best in slot for a given spec? Of course. Will you still need to understand the mechanics to make the best choice for your character? We think so. One of our overall design philosophies for WoW (and Blizzard) is to be simple but have a lot of depth. Some people mistake needless complexity for depth.

Meanwhile, we think there is a lot of benefit of getting rid of problematic stats like defense and accomplishing things like letting casters improve by focusing on throughput over mana pool.

Q u o t e:
In sum, I realized that the guide I was in the process of writing was a relic from a different design philosophy. My initial impression that this was a characteristic of the DK class evaporated after getting my Paladin and Warrior to 80: by simplifying the tank stat game, stat decisions were no longer as simple as making the "best choice". Context had become so important that the person best qualified to make a decision was the player themselves (when armed with a little information).

The community reliance on third-party generated information continued, and with it a renewed investiture in the authority of Wall of Text and Lots of Numbers. A particularly perverse event sticks out in my mind: arguing with an inexperienced Paladin about their stat and gameplay decisions. This player was a disastrous tank: bad threat management, problematic rotations, poor positioning, etc. They defended their decisions with a forum post, a guide incorporating entire text blocks from my original. Here I was, an MT trying to help an OT not suck, and my voice couldn't be heard over the authority of a forum post plagiarizing my own forum post. The seductive certainty of a well-respected poster superceded the in-game feedback of "some bozo" who happened to be the poster in question.

The community needs players confident in their ability to make decisions for themselves, to view their in-game experiences as primary, rather than tertiary, data. Many players feel overwhelmed by the current statistical complexity of the game, as evidenced by their reliance on the presentation of sophisticated argument from abstraction.

Very well said.