Damage meters have historicaly been one of the first addons developed for new MMOs. This makes perfect sense, really. The essence of the fantasy that pulls us into MMOs is 1. I am powerful, and 2. I am special. Damage meters help us to measure just how powerful and how special we are. In the past, changes to dps specs have often come about because discrepencies of damage per second have come to light through dmg meters/logs, and the discrepency has been brought to the attention of Blizzard, who in turn buffs/nerfs appropriately to give us a more balanced raid environment. But how much do they really tell us?
Teh Dee Pee Ess
When we’re talking about measuring the performance of a dps player with a damage meter or log, this works perfectly. After all, aside from staying alive, damage is your one and only job as a dps spec. So damage done is a pretty solid measure of ones performance on a given fight. Sure, there are a few possible exceptions; “Hey, that moron isn’t switching to the LOA!!! SHIV HIM IN THE THIGH!!” “Hey, he’s multidotting on a health reset council fight!! Fart in a jar and mail it to him!!”. But by and large, damage is a good indicator of ones performance level. For tanks and healers though, this is not nearly so cut and dry. Let’s start with tanks.
I R TNAK
Tanks have four jobs really;
- Stay alive – This is an easy one. If you’re tank keeps dieing, that’s a problem. Obviously, there could be a lot of reasons for this; your healers could be failing, your tank could be underusing CDs or not using CDs, your tank could be firestanding. A lot of this can be shown on the damage taken by spell category. This shows a lot about how well the tank is doing. But to compare your MT to your OT is going to give a skewed picture of what’s really going on because the MT probably will have more time on the boss and therefore have more damage taken by a given ability. Here’s where a damage taken per second metric would be useful. I’ve never seen one though. An easy solution is instead of comparing MT to OT, compare your MT to a similar raidgroups MT for the fight. They will likely have similar uptime on the boss.
- Keep aggro – This one’s pretty easy too. If dps are dieing to boss melee attacks, your tank is failing here. This isn’t always completely his fault. Some tank classes don’t have great snap aggro. So a dps that goes full on bloodlust-battlerager-zerker mode immediately may find themself getting instagibbed and losing the fight early. Not really a tank issue. But back to the subject, not something a meter can really show you.
- Position the boss correctly – This one requires the use of ones eyes as well and can’t be monitored through the meter/log.
- Do dps – Here we go! This one is all meter. But with the way vengeance works, tank dps can really be all over the place and we still may have no aggro problems at all. Dps for a tank is largely more of a bonus and not a direct true measure of tank skill.
All in all, the meter/log can show you a good bit about how well a tank is performing. Through damage taken by spell, and comparison with other tanks, you can see how good a tank is at surviving. Can you tell if one class is less desirable than another through this method. Yes! Look at an MT of one tank class from a similar guild with similar gear and compare to an MT of another class, similar guild/gear. How much damage did they take from the same boss? That’s a good indicator. You’ll need to look at several to make sure there isn’t a significant skill difference blurring the results, and even compare on raidbots for aggregate performance using thousands of logs.
WTB HEALZ
Healers are in kind of a different place then dps or tanks though when it comes to measuring performance. This is because our job is really about saving players from death more than anything else. One healer may be knocking the ceiling off the healing done meter, but if they’re not catching low health players before they die, then they’re not doing their job. Also, if they waste spells in such a way that they oom, leaving them without mana when they need it. Additionally, the healing done metric is easy to manipulate to make one healer look better than another, to the detriment of the raid. For example; if I time a large cooldown for just before another healer is set to use their CD, they will waste their CD and I will be much higher on the healing done meter. Having played with players who do things like this in the past, I can tell you, it is exasperating to know that CDs that could be used to win the fight are being sacrificed in this fashion on the altar of epeen. But does all of this mean that the meters don’t matter? Can we gather no useful information about healer performance from HPS or Healing Done? ….Ok, so the way I phrased that, I’m not sure if I should answer yes or no…… Of course meters matter!!! Just not in the cut and dry way we would prefer. You can’t measure which healers are saving lives over padding the meter. It’s just not something you can know from a meter. A good way to determine this is to two heal with a healer you’re trying to evaluate. If folks are dieing, that may indicate that they’re not prioritizing well or are making some other kinds of bad judgement calls.
What about an overall healing spec’s performance? This is where World of Logs is really useful. When looking at thousands of logs from players of all skill/gear/progression levels, a lot of the inconsistent ‘wrenches in the works’ like player skill and raid comp get ironed out with that much data. This is because you are likely to have hundreds of other players with nearly identical skill and raid comp. Statistical outliers get mellowed if you add enough data. This is why scientific studies use a lot of subjects, not just a few. The more people you study, the less impact the wierd cases have on the results. World of Logs has a ton of data! That’s good. It means the average results are a pretty accurate indication of what a given spec is capable of.
What about absorbs? Quit peaking at my notes! Absorbs do have a large impact on the throughput of more direct throughput healers like shaman, h-priests, druids, and monks. They effectively put a ceiling on our throughput by preventing healable damage from occuring. But is it really relavent? The raid environment we are now in leaves the raid’s health pool with either everyone near full health, or everyone below 50% briefly before a raid CD is used and everyone is back at full health. Absorbs extend the ‘full health’ part of that cycle. Meaning direct throughput healers have little to do; effectively getting presniped™. Basically, if you’re not a disc priest or holy pally, you’re going to look a bit worse than you actually are on WoL, assuming you’re running with either a Disc or an H-Pal in your raid. Be sure to factor that in.
But is the overall performance of the spec represented accurately, relatively? Yes. That’s the area where you can accurately make some determinations. So it is safe to say that if a spec is underperforming in HPS by a considerable amount, than yes! There is a problem. They are not putting out as much healing as other specs. With as much data as is available, all variables like raid comp, skill, gear, progression level, do not make a meaningful impact on the results. The data can be interpretted as-is when looking at overall spec’s power level (OVER 9000!!1!1!!!).
So why am I hearing on the forums that HPS doesn’t matter in regard to whether a spec is underperforming and in need of buffs/fixes? Well that’s an easy one. It goes back to our two prime drives for playing an mmo. The idea that 1. I am powerful and 2. I am special. When one’s spec is very powerful or arguably over-powered, the last thing they want is another spec to get buffed up to be competative. Than they would not be as powerful or as special, relatively. Sure, it’s true that HPS is not a completely all-encompassing measure for what makes an effective healer, as we discussed earlier. Nor is it an accurate measure of whether a spec is keeping the raid alive. But it is a fairly accurate measure of a specs overall power and potential. To dismiss it completely tips their hand, illustrating that they truly have no interest in balanced gameplay and merely want to preserve their advantage. If their spec were suddenly at the bottom of the HPS dog-pile, I am confident we would observe a stunning reversal of their opinion. Had HPS been dismissed in 5.1 resto druids would not have received the buff that they deservedly received in 5.2. In fact, to dismiss HPS completely is to say that any healer classes who have ever been buffed based on HPS underperformance (read: all of them), did not need a buff at all. It’s called willful ignorance and it’s en vogue right now.
>luvbacon<
Totem Ploppin' 