Thursday, January 10, 2008

Blizzard's Unintended Consequences

Blizzard is the undisputed king of polish in the MMO landscape but they sure seem to suffer from short sightedness in their game mechanics sometimes. Maybe this is not because of any fault of their own but rather the size of the game they manage. Its hard to predict the effects of a patch on a virtual world when it has over 8 millions random variables running around. A high amount of players definitely seems to increase the odds of someone finding a way to exploit any new changes. Human nature and competition makes sure knowledge of the exploit spreads quickly. Lets look at some recent unintended consequences Blizzard created and ignore the classics like Hakkar's Curse of Blood incident.


AFK Reporting Feature

The Why?
Alterac Valley was suppose to be the ultimate PvP experience in World of Warcraft but has never quite lived up to its imagined glory. After two zone revamps Blizzard decided that it was impossible to balance since so many players were using methods to stay AFK in the starting caves. This was more a problem on the Horde side then the Alliance though make no doubts but sides had plenty of slackers.

The Unintended Consequence
The reporting feature was suppose to prevent players from being able to farm honor by sitting around AFK in the battlegrounds. In Alterac Valley it initially made the odds more fair but within a few weeks people had figured out places on the map where they could go AFK and still look like they were in battle. Even worse the newest form of Alterac Valley gave little honor to the loosing side. Originally thought to discourage being AFK all it did was force them into all the other battlegrounds.

The Causes
1) Blizzard should have know that making Arena Season 1 rewards available for honor would make people looking for easy ways to earn honor. 2) They might have guess that the reporting tool was only effective in long games because the amount of time it took the tool to flag someone AFK. 3) And they definitely should have known that victories in Alterac Valley were usually very one sided and it might not be the best thing in the world to make it so the losing side gained little honor. Add all three factors together and you now have the situation today with even more AFKers spread across all battlegrounds.


Required Arena Ratings and Smurfing

The Why?
Blizzard noticed the trend of raiders only participating in arena games so they could save up points for epic quality weapons. It was much easier to get weapons from arenas then raiding because of the lack of a weapon token system in PvE content. To prevent this trend Blizzard put a rating requirement on season 3 weapons when they came out.

The Unintended Consequence
People Starting selling spots on high rated arena teams. Blizzard created a personal arena rating in season 3 to prevent this practice but it soon became apparent there was already another exploit in use. This exploit was called Smurfing and involved players using multiple dummy teams to feed a single team easy victories late at night. A lot of the top winners of Arena Season 2 were caught using this method which of course meant that most of Blizzard's development on the rating requirement was negated.

The Causes
The arena rating problem has been caused by a single factor and its one the developers seem unwilling to recognize. Buying arena spots, creating a million smurf teams, and moving characters around to each of them is still easier then winning a roll on a good epic weapon. If a token system was introduced for weapons or perhaps the raid loot adjusted to drop more weapons it could have fix this problem. But instead the developers choose to focus only on the PvP side of the equation to achieve a balance.


Conclusion

Now why I may have been a little harsh on Blizzard, I was purposefully highlighting features where they had made mistakes. Traditionally, the company's commitment to class balance and dungeon design has been very high with a lot less mistakes then their competitors. Their good record on the raiding and combat side is in part due to their talent pool from classic Everquest. However, if there is one thing Blizzard needs to work on its reacting to symptoms of a problem instead of tracking down the source.

The introduction of daily quests and the revamp of elixirs was a great example of Blizzard attacking the source of why people bought gold instead of going after the symptoms. However, it took them quite awhile to figure it out since it wasn't in their area of expertise. World of Warcraft's PvP systems are in a similar situation where they have been tinkered with for years but have yet to provide that perfect experience. Still they keep getting better and this has had the unintended consequence of taking people away from raiding. Now blunders on the PvP side are very dangerous since they could make it too easy to get rewards or too hard. If its too easy then WoW risks losing more raiders and if its too hard they risk feeding Warhammer.

11 comments:

Anonymous said...

For the sake of accuracy, locking weapons behind ratings was introduced with Season 3 not 2.

As for the statement that the losing side gets little honour off AV, that's true in practice because people don't really bother to actually play the BG. During the first couple of weeks after the rules changed with 2.3, and people put in the required effort to defend their towers while methodically taking out the opponent's, the losing side could often end up with more bonus honour than the winner rushing the general.

Since then, though, the QQ notion prevails and people give up before it begins, leading to, indeed, very lopsided scores.

Relmstein said...

I'm still not sure if its the QQ mindset or just bad design causing people to loose heart. One of the dumbest ideas was to grant honor for actually killing opponents in a game that requires flag guarding. This guarantees more honor for those on the offensive which is the core problem with every battleground. There are a lot of ideas to fix this problem:

Battlegrounds chould get rid of all honor for kills and double the amount for flag objectives.

Or they could work on a more offensive manner of keeping score rather then depending on flags.

I've even heard one person suggest putting weapons at flag nodes that would let people defende them. Something like a knockback balista or cannon.

Gharr said...

I've seen fights where the losing teams walks out with 60 or less bonus honor. Because the winning team put up a balanced defense and offense kept doing it's job. Those matches are longer than the rush to the end matches though.

Sometimes much longer, and after the first 30minutes of the match, people start asking why there is still any defense, and start berating the defense.

I always felt the point of a competition like this battleground is to defeat the opponent. Not walk hand in hand with them and random who gets the king at the end (so that both sides get nearly the same amount of honor).

I have to wonder how long it'll take the losing side to learn to work together like the matches that have a solid defense and just punch through them. 10 People can't keep 25-30 back. Unless the 25-30 trickle in or stop to actually fight.

Personally I like that the winning team can and does devastate the losing team when they losing team doesn't work together. Why not do to the enemy in this BG what you can do in the others, lock them out of the bonus honor they crave so much. It is a competition between two sides.

heartlessgamer said...

I've always said that honor should not be given for battleground kills. It simply highlights a "me me me" attitude and team-based objectives are forgotten wastes of time.

It is the same argument I've had for nearly every single online FPS game I've played where the most points are rewarded for killing players instead of completing objectives.

I always found it funny while playing Battlefield 2 that the top scoring players usually had NOTHING to do with whether the team actually won. In reality, it didn't matter if a team won, it mattered if YOU scored a lot of points. It is just a contradiction of a design. I can't believe designers sit down in a meeting and say: "Hey, let's make this great team and objective based game, but give scores based on how many players you killed. Oh, but we'll give a couple points to the players that complete the objectives, just to make sure the matches end eventually."

Do I have a solution for this problem? Fuck no. I don't get paid six figures and I don't have the time to figure it out. I have an idea of what I would like to see.

I tend to lean towards an all or nothing rewards system. If your team wins, you get a full portion of the winnings. If your team loses, you get nothing.

Of course, that doesn't work, because most players can't stand losing and it promotes AFK behavior by players just looking for a free ride.

For World of Warcraft, Blizzard just needs to stop trying to save AV and spend some time revamping the other battlegrounds, which are IMHO in far far far worse shape.

Anonymous said...

They really really need to switch to a trinket system for raids. Some raids and five mans have them, but it needs to be implemented in all of them.

What made Blizzard think that Joe Average player really wants to run the same raid 20+ times just to get five or six pieces of loot? Life is too short.

Anonymous said...

Part of the problem is the aging slate of battlegrounds. It is undesirable to have too many, as that would lead to longer queues, but keeping the same 3-4 stale BGs for years is unsatisfying to players.

They should produce several more BGs, and rotate them, so that only 4-5 are active at any given time. Each week, the slate would change at random or on a rotation (more likely).

Anonymous said...

Just curious how you know that the AFK'ers were more common on the Horde side?

Relmstein said...

I played both horde and alliance to level 60 before the expansion. When the battlegroups were first introduced I got lucky since my alliance and horde servers were in the same one. This enable me to try the battlegrounds from both sides and make comparisons.

Anonymous said...

I recently posted a suggestion on the forums for a drop/badge system combo for raids ; feedback/thoughts/bumps appreciated : http://forums.worldofwarcraft.com/thread.html?topicId=4028100889&postId=41178643985&sid=1#0

Anonymous said...

I think the only way to really figure out player reaction is to have people play on the test server as their main server. Or maybe a whole battlegroup would be a test group of servers where changes could be introduced before rolling them out to the rest of the population. Since truthfully the only way players will exhibit certain behaviors is when they have a vested interest. Players who merely copy their character over to the test server are just looking around but have no interest in the outcome (on that copied avatar). They won't find ways to maximize reward/time unless they are going to actually get those rewards and use them permanently. As a consequence, test servers that are temporary can never reveal that sort of behavior.

This is similar to the dilemma I've seen in IT on whether to provide "real" data as test data for systems or made up data (for instance an HR or Compensation application). When it comes to usability, real data always reveals more problems than the test data, because only then people actually pay attention and think of their real needs, they feel like they're working on their own data not playing around or testing.

Relmstein said...

I hate to say it but while Blizzard also needs more testers they could also use some people who have experitise in areas not usually associated with MMOs.

Economists for predicting gold and gear inflation and setting gold sinks without influencing the demand to buy gold.

FPS/Console Experts for making pvp objectives which aren't boring and actually have people excited to play instead of sitting AFK.