Balance analysis of 2.1 WAR proposed changes

[Gamemako]

Lots of rage going around about the proposed changes Mr. Yoshida put forth. To put a bit more reason into the debate, I've run the numbers for a few key situations and metrics.

First issue with WAR was the eHP discrepancy. WAR received no change to base HP, instead rolling up a pair of new eHP-increasing buffs. With the buffs, the eHP situation would look like this:



PLD are in blue, WAR are in orange. Standard eHP stuff: no random effects count, only perfectly-reliable bonuses. There are no differences in defense for which to account, and I accidentally left my crystal ball on the other forums and can't tell you the relative value of SP and RoH debuffs.

WAR was pathetic in this category, but now is actually able to take a punch. Nobody is really running away with it, though -- WAR's advantage at the top end is meaningless (4% more on top of ridiculous). WAR does, however, have a lot of options here with shorter cooldowns.

Next thing is burst mitigation. I've included two basic scenarios: standard burst ability use and peak burst ability. The former is your basic use to reduce damage from groups of mobs or take a predictable spot from a boss. Represented here is Rampart for PLD and 2 uses of Inner Beast (via Infuriate) for WAR. The second is your peak ability use. This is used to survive occasions where a healer went on smoke break and Titan decided to you use as a pinata. For PLD, this is Rampart, Sentinel, Bloodbath, and Convalescence; for WAR, this is Berserk, Bloodbath, Infuriate, Vengeance, Inner Beast, Thrill of Battle, Second Wind, and Convalescence. I have excluded both Hallowed Ground and Holmgang for obvious reasons. The selected intervals are 10 seconds and 20 seconds, comparing both the intended duration for PLD and WAR abilities. Note that a 20-second interval is particularly troublesome for WAR -- I had to create a rotation which had 6.5 seconds of downtime for Inner Beast. It is unavoidable, even in burst situations. Each of the situations includes both the baseline stance bonus in addition to the abilities themselves. For the sake of simplicity, I have neglected the time required to build Wrath for each initial use of Inner Beast -- because the ability is your burst mitigation, you should have it on hand for most situations. If you don't, it's because you're just entering an encounter or you're bad.

First, your humdrum CD drop:





Note that, due to the short duration on Inner Beast, the ability falls behind for longer intervals. This is, as I previously mentioned, unavoidable as a practical matter. However, over a shorter interval, Inner Beast is identical to Rampart with an additional self-heal on top. This self-heal becomes trivial at higher damage rates, which should surprise all of no-one. I included two possibilities for WAR as I did with my previous thread; the difference is now very small due to the strength-dependent portion being reduced by two thirds. Nothing here suggests that WAR is replacing PLD.

Next, peak mitigation.





Same story. WAR scales inversely with incoming damage due to the unscaled self-healing aspect. Unlike the 20-second example from before, the loss of Sentinel helps account for the wonky 20-second rotation of WAR. The extremely-negative scaling early on with the 10-second interval is because healing exceeds 100% of damage, so there is zero additional benefit from Defiance or Vengeance until that self-healing is exceeded. WAR fares well here, but has no advantage at high damage rates -- it's all self-healing (remember that IB, ToB, SW, and Bloodbath are all still significant self-heals).

Finally, we consider the total effect of mitigation effects on the net MP requirement for healers. This is the time-weighted effect of all the aforementioned buffs.



Still the same. But there's something ostensibly missing from this. Considering the overall time-weighted effects of abilities with 3-minute cooldowns is plenty sufficient interval over which blocking has an appreciable and meaningful effect. Including that will look something like this:



Paladin will have an advantage over time any time blocking comes into play. In fact, if you included blocking in other intervals (such as a 20-second interval where 6 mobs are throwing themselves haphazardly at your shield), you'd get pretty much the same result: Paladin has an advantage whenever blocking can come into play.

Now that the data is out of the way, I'm going to force-feed you the obvious conclusions you should have drawn. WAR and PLD end up in pretty much the same place. Regardless of interval or type of mitigation, the two are pretty close, especially compared to where they were. WAR is a bit behind in long-term mitigation due to the lack of blocking. This is not likely to be made up by anything and will continue to be a strength of PLD. How they compare is really a matter of implementation.

Any occasion where blocking can have an appreciable effect, PLD wins out. PLD also sees an advantage where the encounter damage interval is longer. This means groups of trash mobs (e.g. WP farming) as well as high sustained damage (e.g. turn 4) are going to be PLD's strong points. PLD will also see an especially large advantage in AoE mitigation due to its AoE blind effect, which is not included here. For burst-heavy encounters, where autoattacks are little and big hits reliably, WAR has an answer to just about everything. WAR has a lot more cooldowns, making it much easier to avoid being caught with your subligar down. All told, WAR will need to do so in order to be the better tank, but it can be.

However, there's just not going to be a whole lot of content that one can do well and the other can't because they're just too damn similar. Everything about them is similar, from buff-stacking to eHP to where they buy their trousers and tin cans and tin-can trousers. This is about as balanced as you're going to get, for better or worse.