Site icon Intellectuals Bi-Quarterly

Tricks, Logs, and Parses: A WoW Conundrum

Background

A new expansion is on the horizon for Classic WoW. For warcraftlogs.com, the premier raid stats website, this means updating their systems to be ready to track, interpret, and graph all the new spells and encounters. Warcraftlogs is an important tool for competitive players to analyze and improve their raid. For individual players, it also provides a fun digestible statistic that players refer to as their “parse”: a comparison of their dps to all other players’ dps presented as a percentile. Parse a 77 and you’ve done more damage on that boss than 77% of players. Parse a 95, better than 95% of players. Player obsession with parsing can become toxic, but for many it provides a fun way to be competitive in an otherwise Player vs Environment setting.

This system is not immune to being gamed or abused. Someone who really wanted to parse well could be given preferential treatment during mechanics. For example, let the Warlock take blue beam during Netherspite, or just make everyone else move away from the chosen star player during Gruul. Group up adds that spawn mid-fight so they can be blown up by warlock aoe via Seed spam. That sort of abuse is unlikely to go away, and it is hard for warcraftlogs to discretely separate that sort of increase to dps performance from the non-gamed dps performance. Still, it benefits competitive integrity for warcraftlogs to limit player ability to game this system when they can.

When abuse exists it can elevate players a level above everyone else not engaging in that behavior. For top competitors, it can become almost a requirement to compete. And realistically, you could be an extremely skilled player with excellent gear and never be able to compete with these players. A raid might only be able to support one person receiving that kind of treatment, and unless you’re a guild officer you probably won’t have the sway to convince your guild to boost you this way.  

So what is Happening this Expansion?

Warcraft logs is considering changing the way it handles certain types of player buffs. This is driven by two things:

  1. New powerful class buffs provide more ways for abuse to happen
  2. These buffs are straightforward and their effects easier to track by warcraftlogs

Specifically, two new spells in particular are on their radar: Rogue’s [Tricks of the Trade], and Death Knight’s [Hysteria]. There are other additions, but they are more minor. For this discussion, we’ll confine our thoughts to Tricks, because the arguments are the same for Hysteria. The question is how to handle who gets credit for the +15% damage provided by the rogue during the 15s duration of the buff. It’s fairly easy for the logs to separate out the extra damage, if they want to, since it’s just a flat +15% for the duration. There are basically three methods considered for handling this.

Method 1: Attribute the damage to the player receiving the buff

This is current logic pre-expansion, and is the assumed default if a decision isn’t made. Player damage is player damage. If a player would do 100 DPS, and Tricks makes them do 115 DPS, they get credit for 115 DPS. No bullshit.

It’s not too hard to imagine a scenario where this gets abused by the playerbase. Imagine a guild leader who brings 4 rogues to chain Tricks on himself in order to get permanent 100% uptime on the buff. Or, if we want to go less extreme, rogues could get constant whispers from their teammates asking to be their Tricks target. Whispers offering to pay on loot, or pay the gold. Or paypal them. It’s probably already true that the more popular, rich players have unstated advantages already, but we’d prefer to not have a system explicitly tuned to them more than we have to, eh?

Method 2: Attribute the damage to the player giving the buff

In this case, the extra damage done by the player is tacked on to the damage sheet of the rogue who Tricks’ed them.  This is a bit better. It does incentivize rogues to Tricks the highest DPS teammate, in order to get higher raw DPS returns. That isn’t exactly a bad thing, since that behavior ultimately helps the raid.

One consequence of this method is it leads to situations where if one class has a way to cheese a boss, then rogues also have a way to cheese that boss, since they can Tricks the player getting artificially high numbers and copy some for themselves. This method leaves in an incentive for cheesy player behavior, but it’s the same cheesy behavior that players will already try to abuse anyway, Tricks or not, so at least there’s no additional damage created.

Method 3: Attribute the damage to no one

This way attempts to sidestep the issue altogether by not counting the extra Tricks damage as part of player parses. It is still tracked within warcraft logs for analytic purposes, but won’t affect player competition amongst each other.

Immediately this feels like the least complicated option, but presents two quirks. First, it feels bad to have your new move not be tracked in the dps rankings. Second, more importantly, it creates an odd situation where the Rogue may not want to cast it anymore. Tricks costs resources (energy) to cast, plus a global cooldown. If it doesn’t count toward the parse, pressing Tricks is actually a dps loss for the Rogue. It incentivizes this degenerate behavior where the optimal behavior for a Rogue who wants to parse is suboptimal behavior for the raid as a whole.

The best option is probably the one that is least prone to abuse and causes the least disruption to the raid. That is probably option 3, but I think method 2 or 3 are both better than method 1. We’ll see what they decide in the weeks to come.

Honorable Mention: Warriors

Warriors fuck this up because of the way they generate resources. Rage gains increase proportional to damage dealt. So when a warrior get Tricks, even if warcraftlogs skims off the extra +15% damage those warriors still gain extra resource they wouldn’t otherwise have, which translates to more damage. Basically, warriors will always get some version of method 1, at least a little bit.

Thanks for reading.

Exit mobile version