Procrastination Amplification: Punditry on MMOs and games in general.

We Need to be Able to Make Mistakes

Blizzard poster Eyonix wrote about the stat changes in World of Warcraft’s upcoming expansion Cataclysm the other day. The blogging response has mostly been positive so far, Hatch highlights some good changes and Tobold tells us how the removal of defense is a good thing. Alas, Tobold uses an absolutely horrid analogy to make his point. He introduces an imaginary “groupability” stat whose (very negative) description fits defense perfectly. He then goes on to say something along the lines of “see, you wouldn’t want that, and defense is just the same” which is very effective as an argument. Shame though, that I could easily make up a very similar argument for various other stats in WoW – such as stamina for example.

But that’s not my point here, today I’ll talk about why games should have “bad” stats and about what’s good about poor old defense.

I’ve preached over and over again that games thrive on meaningful decisions. If you make stats as simple and intuitive as advocated by Spinks among others, you take away our ability to make bad choices. Sure, wearing intellect cloth as a fury warrior will still be a bad idea, but aside from the really dumb decisions there will be no bad ones to be made. Note that I’m not saying (because I don’t know either way) that Blizzard’s changes already take us to that place, but they definitely take us into that direction and it seems to be what a lot of people want. I’ve read quite a few times that people want to simply look at a piece of gear and know whether it is an upgrade for them or not. I ask those people, why have gear in the game in the first place then? Why not have a single power stat that improves whenever you kill a boss and determines your character’s success in whatever you do?

A thoroughly bad card in the eyes of an experienced MtG player. A lesson to learn for someone new at the game.

Let’s take a look at trading card games, in this case Magic the Gathering because I know that one best. Mark Rosewater, one of the game’s lead designers back when I played, used to tell the players over and over again that the game needed to have bad cards in order for people to make the right decisions and learn. If every card you got had the same power as every other card, then the decision which cards to bring to the table would be pointless. There is a learning process involved in picking the cards you want to play with and it would give players something to be good at.

I don’t know about all you socializers, explorers, and griefers, but as an achiever I want to be able to be good at something. In order to have the potential to be good, however, I must also have the potential to be bad. If you take away all the bad choices and leave only the good ones, then nobody can ever get better because they are automatically making the right choices to begin with.

This brings us back to the defense stat that people seem to be so happy to get rid of. The management of your defense value was one of the aspects that allowed you to become a better tank. The completely unknowing tank might not pick up enough defense at all and would be critically hit by bosses, greatly reducing her success as a tank. That is absolutely fine, as it gives the poor gal the chance to find out what’s going wrong and start picking up some defense gear. Now, some players might be stuck at that point of the learning curve and simply pick up any defense gear they can get, while others would quickly find out that there is an optimum amount of defense to have and pick their gear around that. You might get that cool new breastplate from the new boss you just killed, but equipping it might still be a meaningful choice because you have to take care of your level of defense. The new breastplate might have either too much or too little defense on it, both of which require compensation. Take that away, make the new item strictly superior to the old one, and there is no choice involved whatsoever. That is bad.

Quite a few of the complaints I heard were about how defense has a hidden cap and how it required outside sources to realize that. For one, defense doesn’t have a cap, just a point at which it becomes less effective than other stats – which is true for almost any stat in World of Warcraft. If you have too little stamina  you’ll die, if you have too much, additional points will be useless. If you have a lot of critical strike rating as a damage dealer, getting some other damage increasing stat will be better, if you don’t, getting more critical strikes will be beneficial. The difference between those stats and defense is that defense has pretty much a fixed number that needs to be reached, while most others do not. (Hit rating has the same issue by the way, but that is not being changed or complained about.)

Hydross the Unstable. No, the re-using of art assets didn't start with Crusader's Coliseum. (image: wowwiki)

Could defense rating be a bit more transparent, especially where immunity to critical strikes is concerned? Sure. The effects of defense rating could also vary a bit from boss to boss and maybe depend on other things too, making defense a more interesting stat instead of the “get to 540 and stay there” mentality we have these days. Should it be removed? I think not. I fondly remember designing a set of gear that would allow me to have a high nature resistance while still stying immune to critical strikes because Hydross dealt nature damage that could critically hit. Other bosses wouldn’t crit with their elemental attacks at all, so I could easily drop my defense gear there.

Meaningful choices make or break games. If your game has parts that do not involve the making of meaningful choices, then those parts are as much fluff as graphics and non-combat pets. They can surely contribute to the game’s flair, but they don’t add any gameplay. Raiding already doesn’t have character advancement through levels and now people wish to remove items from the equation as well. What you will be left with is an action game, and a bad one at that.

  • I would argue that there are plenty of bad decisions to be made in play (and if you run enough pick up groups, you’ll see that most of them are independent of gear and spec 🙂 ).

    But is it really necessary for really really bad choices to be present, just so that people who look up their gear on spreadsheets can feel superior? I’ve played fun games where the range of options didn’t allow you to really fail that badly, and I don’t think it affected the gameplay. (Dragon Age, for example is not a heavily gear based game.)

    You seem to be arguing for a definition of ‘a meaningful choice’ which means that there is one definably optimal choice and all the others are shades of worse (shading down to really really bad). That’s not my definition of a meaningful choice. I’d prefer a choice where there were lots of different viable options. If an option is non-viable then it doesn’t need to be in the game, because it simply doesn’t present a meaningful choice at all. And if people choose the non-viable, then it means the game is flawed because the game should provide enough information that no average player would make a really really suboptimal choice. So if you make your optimal choice, it just means you are average. That doesn’t seem like much of an achievement to me. And that’s because there was no meaningful choice involved.

  • Not exactly. I love it when there are multiple viable options available in a game where none is clearly superior to the others. For a choice to be meaningful, there needs to be a variation in outcome, however. If I choose to side with the werewolves in Dragon Age and eradicate the elves, that isn’t strictly better or worse than the other options, but it has an effect on the rest of the game. If, on the other hand, I have a conversation in Mass Effect 2 that gives me two options that both lead to the exact same response, that choice isn’t meaningful.

    The better/worse issue comes into play in two spots. For one, very mathy mechanics make it hard for outcomes to be different but equal. e.g. to make the choice between haste and crit meaningful for a mage in WoW there pretty much must be a difference in net DPS. If both dealt the exact same amount of DPS, the choice would be meaningless.

    I still believe that bad choices need to exist in any interesting environment to allow for good choices to exist. I’m not saying non-viable, because that is an odd concept. It is viable to level your hunter to 80 in spellpower cloth. It’s a pretty bad decision, but you can still play the game. If you have a shred of brain you will start learning from that and realize that maybe some agility would suit you better. As a player (completely free from superiority over others) you get better. Raph Koster (among others) has argued that fun is learning and learning is fun and I think he’s right. If you are automatically at the optimum state, how can you ever learn?

    Here’s Rosewater’s post on why bad cards need to exist, as mentioned in the article. I think much of it applies to other games as well.

    To finish off this extra long comment, yes there are other choices in WoW than gear and of course you can make good games without putting importance on gear. WoW doesn’t have many of the other choices that make Dragon Age good though, and if you take away gear then pretty much the only non-action-oriented choice for a normal raider to make is the distribution of talents. There’s plenty of action-oriented stuff going on, but if that’s what you’re looking for, why not play an action game in the first place?

  • I don’t think defense fits your description, because there were never any meaningful decisions to be made around it. Anyone not capped would be told by even random puggers that he should be defense capped. The only decision you could probably make would be to go beyond the cap instead of switching gems and enchants to stamina.

    I agree that stat choices should matter, and should give you a meaningful choice. I could imagine a future where a tank would have to balance some stat which would increase his threat (via dps) against some stat which decreases his damage taken. But in the current situation defense gives no choice, because not being capped opens you up to the very real risk of being one-shotted as tank.
    .-= Tobold´s last blog ..Being poor in a mature economy =-.

  • Wow, Spinks and now Tobold. I need to disagree with people more 😉

    It is true that defense as it is right now is quite flawed. The decision to go below 540 is clearly bad, though even learning that is a process. (Especially if you try to understand why and not just follow some guideline.) I still like seeing the advance of tanks from “Look i have 560 defense, I am so awesome” to “Look, I have exactly 540 and spent the rest on better stats.”

    Yes, all that can be easily learned by reading EJ, but doesn’t that rather point towards making defense more interesting instead of killing it? How about dual-wielding bosses with higher crit rates but less damage per crit (meaning being crit-immune isn’t vital anymore)?
    There’s also a dire lack of fights like Hydross that actually require you to modify your gear accordingly (and personally) instead of following some generic guidelines.

    As for threat vs mitigation – good point Tobold. In theory that is already in the game. Alas, threat matters so little and mitigation is so important that gearing for threat is rather pointless.