A Simpler Cataclysm

For its third expansion of World of Warcraft, Cataclysm, Blizzard is cutting things out. Numerous stats are being changed or removed. Confusing or redundant attributes are being reconfigured. The game is being simplified, which begs the question: why? In an official thread on the company’s forums, Blizzard said, “Our ultimate goal is make gear a more interesting (and less confusing) choice by making each stat valuable to more players.”

Opinions concerning this are mixed. “Why are they changing things again?” “They’re dumbing down the game!” “Finally!” My opinion falls with the latter. Complexity does not indicate strategic depth; it can obscure attempts at strategy. In order to make a meaningful strategic decision, the player must have a clear view of how it affects his game. A choice from which the consequences are impossible to predict results in no real choice at all, especially in a game as reliant on tweaking your character as WoW.

WoW‘s predecessors, tabletop board games & RPGs, didn’t have computer processors to calculate fine outcomes. The players had to process the mechanics themselves. If they didn’t grasp the system, then they didn’t play it. Understandable rules were a requirement. With the advent of computers, however, more of these mechanics were hidden in the background where the developers could make finer changes to them and keep them from cluttering up the player’s view.

As a result, many attributes did not make sense to the player. Take the Luck stat in computer RPGs like Oblivion. It usually affects critical chances, evade rates, and other nebulous odds. Its precise effect is rarely explained; most players simply have faith that it does something and leave it at that. Now suppose that a player has to choose between a helm that gives +40 Defense and a helm that gives +40 Luck. With such little information on their effects, how do players choose?

Unfortunately, they react the way that humans always have when confronted with the unknown: rituals and stories. The Aztecs sacrificed humans to strengthen their gods. Baseball players don’t change their socks during a winning streak. We focus on 1 or 2 stats exclusively, or make sure they’re all as even as possible, or save certain sets for special occasions with just the barest notion as to why. We worship numbers that we barely understand.

Among the player base, however, there are scientists: people who aren’t content with vagueness. Blessed are the FAQ makers, for they correlate hours of data and reverse-engineer the formulae for public browsing. And some of them are complex – percentiles multiplied by percentiles, diminishing returns, damage reduced by armor, which is in turn reduced by armor penetration, nested fractions so tangled that most people are tempted to call in an algebraic exterminator to gas the program for rogue parentheses. This is the end result of game developers who spent years not caring whether the players understood their mechanics. Most people couldn’t make heads or tails of it, but some people could, and, once they post the solution, then all strategy ends.

“For maximum damage, 280 Hit Rating, 24 Expertise, get Armor Penetration to 20% and then invest everything else in Strength,” reads a typical post on websites like Elitist Jerks.

Players worship these guides. But it isn’t a strategic decision; it’s a math equation that someone else solved for us. Is a game strategic for making its systems so obtuse that you need to ask a gamer scientist for help? That’s like saying that an adventure game is challenging because you have to try every item in your inventory to solve a puzzle. When I think of strategic RPGs, the first one that comes to mind is one of the simplest: Paper Mario.

A kindergartener could understand Paper Mario‘s system. Your hit points never went above the double digits. Damage was a straight Attack – Defense formula, with a bonus if you had good timing on the Action Command. Improving your stats by just a point was cause for celebration. Choosing between a +1 Attack badge and a +1 Defense badge was much harder, though. Both would profoundly change your gameplay, and you knew it. It reminded me of board games like Settlers of Catan and Puerto Rico; they give you simple relationships and then make you sweat over them.

There’s one aspect of WoW where the results of player decisions are clear. Don’t Stand in the Fire: a tense reminder for raiders to watch their feet as well as their health bars. If you see flames, you move out of them; it’s simple, right? But it forces you to choose between casting a spell and avoiding damage. It’s not the only complication in boss fights: fires, exploding players, adds (smaller enemies attacking along with the boss), breath weapons.

These simple concepts spur complex plans, debate, and experiments because the cause and effect are obvious. “Don’t stand in the fire, don’t be near anyone when the timer goes off, spread out in Phase 1, collapse to the right in Phase 2.” If you stand in the fire, you die. If you’re next to others when you blow up, they die. If Blizzard treated boss fight mechanics like they currently treat stats, then the fires would be invisible and only running complex mathematical formulas would allow you to know where the damage was originating. That would be a wonderful challenge! I can already hear the raiders screaming in joy!

Simplicity is visibility, and visibility means understanding. Understanding doesn’t necessarily mean mastery, though. Just because something is made of simple components doesn’t mean it does not allow you to think critically. You can write the complete rules for chess on a single sheet of paper. But critical thinking and strategy requires something more. It requires a variable.

In a strategic game, a plan succeeds or fails by the variables involved. You observe the situation, predict the other variables, and then plan how to succeed. The game only stays interesting if the same plan doesn’t always work. In chess, there is a human opponent that tries a different strategy himself for each game. WoW has numerous bosses with varied attacks and patterns. If the same decision works against each one, it gets boring. Figuring out how to maximize your damage against a target dummy is not interesting; determining how much damage you must sacrifice in order to survive a particular fight is.

And that’s where complexity interferes. Complex systems emphasize process; simple systems emphasize choice. It’s easy to set yourself up in a simple system. Want HP? Damage? Spellpower? Take your pick. Then you’re hit with the tough decisions, tradeoffs between offense and defense, healing others or saving yourself. Did you make a few bad decisions? Then try a few others. Change your strategy and test it; repeat until you find something that works. Have fun.

Strategic flexibility doesn’t come from equipment, though. Gear is a long-term investment. After the first few pieces, you’re lucky to upgrade it once per week. Most gear provides gradual upgrades, instead of making you rethink your strategy. Exceptions, such as trinkets that increase your offense if you sustain a chain of attacks, are few and far between. It’s hard to change and it barely influences your tactics. In short, it’s boring. We need some aspect of the game that can be quickly changed to match different strategies. Luckily, Blizzard already has one.

Talent points and builds differentiate your hero from every other hero. With a little knowledge and some gold, you can make your character play completely different in minutes. That still didn’t provide enough flexibility and choice, though. Blizzard took a step in the right direction when they implemented dual-specs: being able to switch to a second character build between battles. Now in Cataclysm, the designers want to use talent points for tougher choices than +5% critical strike chance. Good for them, I say. Too many talent points were tied up in no-brainer decisions, like minor boosts in armor or spellpower. Few of them unlocked new skills or really changed how you played; they just reinforced what you would already do.

This is where Blizzard’s initial design comes back to haunt WoW. Originally, talent points were given every level. With Cataclysm, endgame characters will have 76 talent points to spend. That’s enough points to fill an entire talent tree and then some. Devoting each of those points to something important is an insane amount of work to balance. So why not reduce them? Trim the talent trees, condense them into important choices, and give the players just 38, 19, or even 10 talent points. It’d be easier for the developers to make new, exciting talents, then set them so that players have to make tougher choices than just “which tree should I put most of my points in?”

But even with just a few talents, they couldn’t be perfectly balanced against each other. Eventually, a gamer scientist would crunch the numbers and declare that the Hunter’s Marksmanship tree regularly causes the most damage against a training dummy. And everyone would suggest switching to Marksmanship, because it works best against the training dummy and probably the actual bosses as well, who are much less likely to let you analyze their stats before they eat you. If you can’t balance everything out perfectly and don’t reveal all the details on your bosses, how do you keep some variety among the endgame players?

Dungeons & Dragons 4th Edition had a similar problem to solve. The mechanics were changed from previous editions to give each class abilities that were similar in power, but diverse in effects. As individual powergaming became moot, a new form of optimization arose: group synergy. People realized that many powers set up opportunities for other players. The fighter could push a goblin into the other enemies, setting up an opportunity for the wizard to unleash an area attack. A warlord’s command power could shift the rogue into a flanking position so that he could backstab. Players quickly began identifying each other’s powers and then picking their own powers to complement them. The same could happen in WoW.

The one variable that a strategy guide can’t predict is your personal group composition. With 10 classes, each of whom can choose between 3 talent specializations, there are more potential combinations for a 10-man raid than there are current WoW subscribers. What if your talent spec not only affected yourself, but everyone in the raid as well? The framework is already laid down. Certain talent bonuses, like the druid’s moonkin form which bestows a +5% critical strike chance, are shared amongst the entire raid. Others have marking abilities, like warriors’ Vigilance, that apply a bonus to one ally and a buff to themselves when it activates. Right now, some talent specs are better at this than others; priests who have spent most of their talent points in the Discipline tree, for example, have abilities like Power Infusion and Pain Suppression. The effectiveness of these powers depends on your raid makeup and whom you choose for the buff. There should be more abilities like these spread across the classes and specs. Doing so would encourage the same group strategizing that boss patterns do. Since their usefulness depends on the raid’s makeup, each player has to think for themselves.

That is the kind of strategic depth that I want in a game. Vague attributes hinder casual analysis, driving players to read online FAQs for the optimal build. That isn’t strategy. Strategy is looking over a boss’s attacks and suggesting different positioning, or discussing which builds would complement your buddies. Strategy is not finding some perfect solution online; strategy is figuring out what would work best against these enemies with these allies. Simple mechanics emphasizes the external, not the internal. It emphasizes why we play the game. We do not play WoW to solve algebra problems.

Jeff Groves wishes his raid didn’t think roasting his pet helps kill bosses.