Thursday, June 08, 2006

Damage Meters

Another week, another MC run. This one didn't go quite as smoothly as the first, probably due to class imbalance. We had about 8 warriors, 8 paladins, and were very low on actual dps classes. We did get our 8 bosses, but we ran about 20 minutes over our scheduled time of 3 hours. We didn't get any paladin loot, but one of our paladins got a [Vendorstrike].

In any case, I got 3 melee DPS in my group, so I had to healbot most of the instance just to keep them alive. Oh well. I did try to get in some melee.

I also downloaded and tried out a DamageMeter for this raid, just to see if I could quantify my performance. I also respecced (yet again!) to 19/32/0. Curiously, I know (and DamageMeter confirmed it) that I do less DPS with this build, but it feels much more satisfying. I guess I really don't like the whole guessing-game that is Seal of Command.

Anyways, I'm not really sure how accurate the stats are. I have no idea if it synchronized itself with other people, or whatever. But I like my results, so I will pretend it's accurate. Here are some stats that I think are probably the important ones for paladins:

Damage Done: #30, 12% of lead
Healing Done: #2, 88% of lead
Curing Done: #4, 58% of lead
Net Healing: #6, 77% of lead
Overheal %: #40, 14%

Now, the position values are not that useful, mostly because who I'm comparing results to changes. But I'll leave them in just for comparison.

I think the numbers show that I healed a lot. 88% of lead is quite good, even if it drops to 77% when you consider that some of that was healing myself. The damage is low, but understandable in light of the healbotting with a 1H healing weapon. The curing amount is pretty weak. I need to be much closer to the top value. I think part of my problem is that Cleanse has a 30 yard range as opposed to the 40 yard heal range, so I am misjudging distances.

The overheal %, or amount of my heals which were wasted on a target with full health, was really good in comparison to the entire raid (including the non-healers somehow). However, 14% seems a bit high, especially as I wasn't healing tanks for a lot of it. I think the raid in general needs to work on our overhealing, and I would like to get my number down to less than 10%.

So basically, I need to do a bit more damage, heal a little bit less, and cleanse a little bit more. Sounds like a plan.

3 comments:

  1. I try not to look at damage meters cause it gets me depressed lol.

    Vendorstrike! LOL. That weapon makes me laugh.

    Priests have benediction/anathema which transforms to fit their role. I wish Paladins had such a weapon. It would transform from a fast 1 hander for tanking into a slow 2 hander for dpsing!

    ReplyDelete
  2. Overhealing when healing tanks is bit different that overhealing on random dps, I think. On tanks, it's a margin of error, but on dps it's just a waste, usually.

    As for damage meters, I find looking at absolute position is pointless. I think that your 20 dps people should be really close together, so you should automatically be in the bottom half. Looking at % of lead is the best way to go. My goal for healing-melee is 40% of lead.

    It was the first time in MC for the paladin who won Vendorstrike, so he was excited. The other paladins were all getting dismayed at the absolute lack of Lawbringer.

    ReplyDelete
  3. Damage Meters system for merging results often counts a lot of damage/healing twice. If you really want accurate stats, use WWS stats. It requires no addon and simply merges together combat logs (recorded by typing /combatlog).

    You can find it here: http://www.lossendil.fr

    If you really want to get accurate, look up the AutoCL addon. It maximizes combat logging range to 100m instead of 30m.

    The resulting statistics are dramatic... average/max with and without crits, % of damage, % overhealing, % overhealing per spell, class summaries. It's really an advantage for any raid who has a leader willing to run it.

    Cheers,
    Aeonix, Shaman, Hellscream (H)
    Telsoa, Paladin, Hellscream (A)

    ReplyDelete