Eric, of Elder Game, published a masterful article yesterday that gives a lot of insight into the MMO development scene and then goes on to apply that industry knowledge to speculate about the state of World of Warcraft.

Here are a few snippets to give you a taste:

On product lifetime cycles

I like the “tedious maintenance stuff.” I actually prefer working on the live team. This makes me very unusual in the MMO industry. I am also a pretty good engineer with a lot of experience, which means I don’t often end up on live teams — too experienced. At Turbine, I had a hard time getting onto the Asheron Call 2’s Live Team, because I was expected to help develop their next generation MMO engine instead. I wanted to work on AC2 after it ships?! None of my managers could understand why I wanted to be demoted like that!

As a software engineer the article rings true; companies - especially larger entities - will develop new products/features using an 'A Team' of rock star developers, and then assign their less valuable and/or newer developers to maintain the final product. The theory is that the A Team will implement a solid base that is then easier for lesser skilled developers to maintain; unfortunately it rarely works out that way for a wide variety of reasons. (Skill, complexity, and vision to name a few.)

On game balance

I found that the Feral Intendant class [in Asheron's Call 2] was 30% overpowered, and that’s why so many people were playing a Feral Intendant. Yet somehow, reducing the power of the Feral Intendant to the correct level did not suddenly make the game more fun… thousands of players were complaining and nobody was telling me they were happy about the change. Weird! I double checked my calculations. They were correct. So what had gone wrong?

Turns out that the people who played the other classes available to that race had taken on an “underdog” mentality. The people who played Claw Bearers liked that they were woefully underpowered compared to Feral Intendants. It was like playing the game on Hard Mode. And the people playing Feral Intendants liked playing on Easy Mode. In balancing the game I had failed to understand the needs of the people playing it. I just ham-handedly fixed the equations, instead of solving the problem with the finesse it needed. It was one of my more serious missteps. (And it’s a great example because I think it’s pretty obvious in hindsight. Most mistakes were much more subtle.)

My experience with game balance is from the point of view of a MUD developer, and I can assure you that it is always a prickly business. The human aspect that Eric brings up is perhaps the most difficult part of the entire equation; players cannot be rationalized with raw numbers, and so it becomes very important to be tuned into all of the foibles of your player base. No spreadsheet can tell you how someone will react to their class being tuned.

On World of Warcraft

But here’s the weird thing: WoW is exhibiting the same symptoms as AC2 did when I was doing the designing. The B team is in charge.

In February, we learned that lead designer (and part-time producer?) Jeff Kaplan had stepped away from WoW, off to work on the next big Blizzard game. However, if you were watching the game before that, it was obvious that major leadership changes had already happened months earlier. My guess is that Jeff Kaplan started moonlighting on the new project long before February. And many of the other key WoW live team people have also switched over, or are working on WoW only part-time.

[...]

When we say that WoW is “polished”, what we mean is that it is surprisingly clean of linty little bugs like these. But that’s changing.

More and more little mistakes have crept into the game recently — changes that are positive on the surface, but have not been implemented with the finesse that makes them worthwhile. Mana expenditure rates have changed, rules for dungeons have been tweaked, the cost of items has fluctuated. It all seems useful. But it’s usually full of little side effects. Worse, it doesn’t take the human equation into account: it doesn’t counter-balance for the actual needs of the players very well. There are ways to meet both goals, but you have to try a lot harder at it than WoW is.

Remember when WoW class balance happened every six to eight months? Players were actually excited when their classes’ turn came around. I remember being so astonished to see players that were actually happy to have their classes redesigned. But now, every class is fiddled with every few weeks. It’s not exciting anymore. Instead of sitting on the changes and carefully honing them, the designers are just firing out every new idea they have, willy nilly, until they get it right. But here’s the thing: it doesn’t matter if you get it right. It matters if players are excited and having fun. Balance changes are happening too fast, and for too little benefit overall.

Although I have no proof of it, this just feels correct. In the run up to Wrath of the Lich King the Blizzard developers started making upcoming class tweaks more and more visible to the player base in advance, and seemed to be relying more and more on community feedback to find their direction.

The massive changes to feral Druid tanking mechanics are an excellent example of the B Team at work - someone got it up their nose that ferals needed to change direction, and completely turned the spec on its head. But not only that, the changes came fast and furious, and feral tanks had to constantly adapt to tanking mechanics and optimal gearing choices that kept shifting beneath their paws. Worst of all, the centerpiece of the new feral tanking system - Savage Defense - was released in a broken and buggy form and as far as I know is still not working as intended.

Final Words
All in all, an extremely insightful article by Eric, and I highly recommend that anyone interested in MMO development go read the rest - it's well worth your time.