Lifetap change reverted

Q u o t e:
1st page.


Big win if it makes it to live untouched, but I'll reserve judgement on 2.4 till it's finalized. Who knows what other nerf they might decide to toss on us to compensate for reverting back lifetap. Hell, they might even consider it a warlock buff, justifying nerfing us somewhere else.

No other warlock nerfs are planned for 2.4. One of the things we look at is class representation in arenas (although normalized for class popularity). In this particular chart, a value of 100% means the class is represented as we'd expect, a value over 100% means the class is represented that much more often than we'd expect, a value below 100% means they're represented less than we'd want (obviously this chart doesn't include a spec breakdown in any way).

2v2 3v3 5v5
Druid 276.0% 184.0% 80.5%
Hunter 43.0% 50.2% 43.0%
Mage 8.7% 96.0% 96.0%
Paladin 19.7% 29.5% 147.4%
Priest 113.3% 164.8% 185.4%
Shaman 37.8% 50.4% 138.6%
Rogue 144.2% 175.1% 61.8%
Warlock 149.2% 93.2% 111.9%
Warrior 130.4% 90.7% 79.3%

Locks are doing pretty well overall, but what had us worried was that it appeared to us a month ago or so as though warlocks were on an uptrend. However, the evidence is strong that that trend is reversing, which means significant nerfs aren't really what we want right now (we really don't know where those numbers will settle down).

Edit: one other detail regarding the chart, this one is set at 2200+ rating.

Q u o t e:
Those numbers are very interesting. How are they normalized, exactly? A few of the data points strike me as unreasonable - for example, the core of many gladiator-level 5s teams is warrior/paladin/priest/shaman. Paladin, priest, and shaman numbers seem correct (highly overrepresented), but the data implies that warriors are underrepresented in high-end 5s. Seems unlikely.

There are lots of warriors. Considering how many there are, fewer of them than one might expect are super-successful in 5v5.

That doesn't necessarily mean we'd buff them for 5v5. We aren't slaves to the charts, they just help inform our decisions. =]

Q u o t e:

This leads the balancing team to the conclusion that the higher concentration is an indication of some sort of class imbalance. Unfortunately, this model on it's own is completely useless as a balancing metric because it fails to take into account the effects of synergy. Some classes just "work well together".

A better metric would be to look at the partners of all of those over-represented classes and attempt to identify the point of synergy. I hate to be the harbinger of nerfs, but how many of those Warlocks, Warriors and Rogues are partnered with Druids?

Synergy definitely plays into the results. In addition, there are also all kinds of class relationships that come into play. For example, in most cases we'd expect that the over-representation of druids in 2v2 accounts somewhat for the underrepresentation of paladins, and probably also contributes to the mage 2v2 effect.

Q u o t e:


it seems pretty silly to weight the participation numbers for very high end pvp by the entire population. High end teams will take the classes that work, not the classes that are popular. A 2200 team is not gonna settle on a player just because they can't find the right class.

While class popularity is likely to mean less the smaller the sample size, we still feel like it's worth looking at (and yes we look at un-normalized stats too).

We definitely feel it's still somewhat relevant at the 2200 mark, there are plenty of anecdotal cases we know of on the dev team where the arena team in question would love to use certain group compositions but isn't able to due to not finding a player of that class with enough skill and gear to make it happen.

Q u o t e:
kalgan shaman would love a heads up on if the flametongue revert with this build is intentional or not.


grats to locks on having their nerf reverted!

Yes the flametongue revert was intentional. After putting it through its paces we felt like we'd gone too far on that one.

Q u o t e:


ouch, there went Push's theory that this was just so you guys could put a workgin version up on the PTR.

Any reason for this revert when a working version was never tested on the PTR? Seems like you guys jumped the gun here.

The PTR doesn't get every build that gets made, there are generally several internal builds in between each PTR build that we're able to test on.

Q u o t e:


I think you missed his point. The normalised representation in the 2200+ bracket is significantly different to the representation in the 1850+ bracket. The latter would give a better overall picture of pvp balance since:
a) It's a much larger sample size
b) It includes non cookie-cutter team comps

Care to post an 1850 chart please?

Here's 1850...

Druid 184.0% 138.0% 92.0%
Hunter 50.2% 50.2% 50.2%
Mage 61.1% 87.3% 87.3%
Paladin 68.8% 88.4% 137.6%
Priest 133.9% 133.9% 154.5%
Shaman 75.6% 88.2% 138.6%
Rogue 154.5% 154.5% 82.4%
Warlock 121.2% 102.5% 102.5%
Warrior 90.7% 85.0% 85.0%

Q u o t e:


All I'm curious of, is this:

Am I to expect this game evolving into World of 3v3craft or World of 5v5craft?

That might depend on what you'd like to believe. I think it's fair to say that we see much better raid representation by different classes/specs than we did before arenas existed. So, I don't really think that the existence of arenas and balance changes that occur as a result of them necessarily imply worse raid balance, or that raid balance doesn't matter.

Q u o t e:


... We get the stigmata of being overpowered ...

I wonder whether that explains the actions of the Romans... Sorry, had to. ;]

As far as hunters go, yes, we do feel as though they're still represented less than we'd like. I think that's safe to say. That being said, we have seen some meaningful improvment.

Q u o t e:
Kalgan: do you balance PvP with specs in consideration, or merely overall representation? Or a bit of both? The whole cookie-cutter phenomenon has taken hold pretty solidly, which I'm not sure can be entirely avoided (something will always be "the most efficient"). However, it seems that in the case of some classes, the cookie cutters are more than just "most efficient"; for example, Water Elemental is seen as absolutely and without a doubt necessary for arenas.

Are factors like this taken into consideration, attempts made to look into nonviable specs and use them as a solution to fixing lowered representation in certain brackets?

We certainly do take it into account. For example, we aren't bamboozled by the numbers into thinking shadow priests or feral druids (for example) are where they need to be. That being said, players can change specs more easily than they can change classes, so we do view class balance as being more important than spec balance (not that spec balance isn't important too).

Q u o t e:


You really forgot to highlight PvP changes that also effect classes in PvE as well.

They almost always affect classes in PvE as well. It's just a question of how much and in what way, and whether we're ok with that. The same is true in reverse for the most part (threat, etc as exceptions).

Q u o t e:
How come the chains on Kalgan's Lich don't spin?

Mike tells me we'd have to charge a few bucks more per month for me to get that upgrade. Can't have it all I guess. =[