So does every hate new PLD now?

[Lyth]

The reason why you don't rebalance numbers immediately after a rework is because there's a learning curve associated with making changes. PLD currently has a much bigger positive skew than the other tanks, which indicates that there's a lot more variation in performance of the bottom 50% of players than the top. Ultimately, that's either going to resolve with time, or it'll persist and indicate that there's actually a problem in the design difficulty.

There isn't a hard binary between 'being good with raid buffs' or not. If two non-buff providers have the same damage over time profile, by definition they have both rDPS parity and identical contributions under raid buffs. People just need to stop equating the latter with aDPS, because that's not what it measures.

[Shurrikhan]

The reason why you don't rebalance numbers immediately after a rework is because there's a learning curve associated with making changes.

Right.

There isn't a hard binary between 'being good with raid buffs' or not.

Also right, but no one had made that claim anyways, only that Paladin was by far and away the job that made the worst use of raid buffs. Which it objectively was.

If two non-buff providers have the same damage over time profile, by definition they have both rDPS parity and identical contributions under raid buffs. People just need to stop equating the latter with aDPS, because that's not what it measures.

This, on the other hand, makes little to no sense.

rDPS parity is not the same as "identical contributions under raid buffs." rDPS is simply a metric that moves 100% of the value created by synergies from the exploiters to that of the buffers. If you are comparing exploiters, it is the worst metric you could look at for measuring parity.

When looking at the most a buffer can bring to an average (if increasingly meta) composition, you look at rDPS, because that's the only metric that accounts for their synergetic value. When looking at the most an exploiter can bring to an average (if increasingly meta) composition, you look at aDPS, because that's the only metric that accounts for their synergetic value.

If two tanks have the same rDPS but one has more aDPS, that means the latter is giving that much more rDPS to his team (even if that job doesn't get credit for that excess itself by that measure); their buffers are getting that much more rDPS.

Or, to put even more simply, between two choices with equal rDPS, every buffer is nerfed slightly (deals less rDPS, which is the metric that rewards buffers for team synergy) when taking the job with lesser aDPS (which is the metric that rewards exploiters for team synergy).

In comparisons between jobs that are purely exploiters, aDPS parity across sufficiently large sample sizes is precisely the closest measure we can get for identical contributions under raid buffs. People should equate "aDPS parity" and "identical contribution under raid buffs", because that aDPS is literally a measure of performance under party-wide raid buffs.

(nDPS, on the other hand, will be made useless for any broad comparison simply due to winner-gets-all skew of single-target buffs. One can't see how large the difference is between the job perceived to be the best target and its runner-up, because only categorical first place gets anything.)

[Lyth]

...

Two tank players with identical un(raid)buffed damage as a function of time curves (i.e. dps(t)) will have identical average rDPS. Also, if you were to pair them both with any raid buff of your choice, they would offer the buff provider with the exact same DPS benefit. This is the very definition of DPS parity. Their aDPS totals will only be the same if their raid composition is also identical and raid buffs are used at the exact same times.

[Shurrikhan]

Two tank players with identical un(raid)buffed damage as a function of time curves (i.e. dps(t)) will have identical average rDPS. Also, if you were to pair them both with any raid buff of your choice, they would offer the buff provider with the exact same DPS benefit. This is the very definition of DPS parity. Their aDPS totals will only be the same if their raid composition is also identical and raid buffs are used at the exact same times.

rDPS on a non-buffer is not influenced by the jobs' damage dynamics (function of time curves) beyond the basic multiplicity between its own buffs and whatever potency falls therein and the timing of the fight; it is, for non-buffers, absolutely irrelevant to team synergies... because the metric was made specifically to be irrelevant to team-synergies for non-buffers.

aDPS on a non-buffer, because it accounts for raid buffs, which in turn care about when that damage happens, is influenced by the jobs' damage dynamics, and is therefore pertinent to team synergies.

Only if two jobs have both the same aDPS and rDPS would they have the same damage dynamics as relevant to raid buffs.

If they have the same aDPS, they will produce the same contribution to the party.

Yes, one may thereafter have lower or higher rDPS and therefore produce a little more or a little less than the other when solo or in light parties, but performance when solo or in light parties is not a distinction that matters to any rewarded challenging content in this game.

Their aDPS totals will only be the same if their raid composition is also identical and raid buffs are used at the exact same times.

You have up >100,000 parses from which to compare, where top percentiles increasingly concentrate towards the very same compositions.

Nor would that be relevant unless one comparator were specifically only taken in raidbuff-light comps on the mere basis of perceived less raid buff value and the other only in raidbuff-heavy comps, yet even top-20 speedruns will quickly show that's not the case for any comparison one attempts among tanks.