Quote Originally Posted by Shurrikhan View Post
It is objectively counting any and all throughput.

Mistakes will have a cost to, at minimum, one's own throughput, but how much so is still quantifiably measurable. The exact mistake -- e.g., that I flubbed a combo while you misaligned your CDs -- is irrelevant; how much they actually impacted our performance is all that will ultimately matter.


Let's pretend for a moment that this wasn't already specifically regarding questions of whose throughput fell short, such as why a group is failing a dps check, or that such a question is somehow not most directly answered through metrics related to throughput.

You seem to be falling under the misconception that parsers can only measure DPS. ACT may call that satisfactory (though it includes more than that as well, just not via its mini display), as it was made only for those who particularly needed DPS metrics (for raids that are ultimately cleared or failed because of DPS), but that's far from from the limit of parsers.

Percent active time, enemy active time against you (a.k.a., kiting percentile), (de)buff uptime, specific (de)buff's uptime, specific (de)buff's uptime against a focus target, avoidable damage taken, number of instances of avoidable damage taken, vulnerability stacks acquired, deaths, raid time spent rezzing you, healer % uptime spent healing your avoidable damage taken, total mitigation, average percent mitigation, average %HP decrease per strike taken, average %HP decrease per strike taken above a set threshold, damage percentiles by source, incoming damage by source, effective healing done, overhealing done, average overhealing percentile, average overhealing cast as portion of total healing casts, damage dealt to specific targets, overkill percentage, and the like are all displayable in real-time through parsers.
Ah, no, I'm not falling under any misconceptions. I specifically denoted how I usually see others focusing on dps.

I see what you mean though; yes, technically, a parser can track a ton of variables. The problem I'm trying to flag stays the same though; it requires someone to go through that volume of information and parse it out into a usable context, which again, frequently devolves into human perspective. Over-selling the "data purity" of a parser doesn't remove humans from the equation, ultimately.

Either way, I see that the discussion is still moot though...

My statements at the start of this entire still hold true to me:
- I can see how a personal score or gauge could be fun
- I personally feel group or public formal parsers are a step in the wrong direction
- So ultimately, the current implementation seems fine as it is and Dev resources could likely be better spent elsewhere