Improved tanking simulator

Just graphed all the hits I took in that fight, and the histogram is as close to a perfect rectangle as you could ask for; no central tendency at all.

Edit: No, it's not; I was telling Excel to do the wrong things.
_________________
Scarlet Crusade - 0/49/12 - 11/11 Kara, 2/2 GL, 4/6 ZA
Healing can be variable. DPS can be variable. My job is to be a constant.

Sorry. The people around here post so much sharp analysis I just assume you're all engineers.

A histogram is a bar graph listing how many times various events occurred. I made one with the damage numbers from all the hits the Curator landed on me, to see if I'd get...

a) A bell curve, meaning you're quite likely to get values in the middle of his damage range and unlikely to get extreme highs or lows. In this case, although the damage might be listed as 2000-4000, most of your hits would fall right around the 3000 mark with the occasional high or low:

And Blizzard's coders take a /bow

Can you order it by time? To see if there's any kind of pattern, maybe? For example, y-axis is the strength of the hit (2000-4000) and x-axis is the time of the hit. Might be interesting to see /shrug
_________________

It's not all that interesting to see, but here it is:

http://www.29tharmor.com/WoW/Curator_01.gif

The 3rd and 4th attacks are the crushes, the two final attacks are with Ardent Defender active. There were three normal hits with HS up, which don't stand out at all on the graph.

Next time I'll get the entire fight and I'll be able to graph my actual health over time, interesting stuff like that. Won't be this weekend, though, 'cause it's Stanton's turn to be MT.
_________________
Scarlet Crusade - 0/49/12 - 11/11 Kara, 2/2 GL, 4/6 ZA
Healing can be variable. DPS can be variable. My job is to be a constant.

We don't have permission to see that file =x
_________________
<3

Ah it's working now
_________________
<3

Please ignore that one. I had some kind of math-incompetence debuff when I did it; I'll produce something better at lunch.

I got some more Curator data yesterday while we were being slaughtered by Astral Flares, as well as a log of our Maiden kill. I'll make those available ASAP.

http://www.29tharmor.com/WoW/Maiden_03.txt

There's a short hole in the log at the beginning, where I wanted to reassure myself the log was on. I typed /combatlog, it said "Okay, it's off" and I had to quickly type it back in to reactivate.

Analysis to follow.
_________________
Scarlet Crusade - 0/49/12 - 11/11 Kara, 2/2 GL, 4/6 ZA
Healing can be variable. DPS can be variable. My job is to be a constant.

Maiden's average hit on me was 2843, after a 285 block and 104 soaked by Blessing of Sacrifice. Hardest normal hit was 3402 and weakest was 2193.

Add 285 and 104 onto those to get the after-armor damage.

My reduction was 56.36%, so divide those numbers by 0.4364, and I get that Maiden's average raw damage is 7399, and her approximate raw damage range is 5917 to 8687.

Her attack rate is nowhere near as neat as the Curator's; she has short casting delays for Holy Ground, Holy Fire and Repentance. She once attacked me twice only 2.032 seconds apart, but a sample I took over a third of the fight suggests her effective attack speed is more like 2.4.

I'll work out the cooldown and damage for Repentance and Holy Ground later - right now, lunch break just ended.
_________________
Scarlet Crusade - 0/49/12 - 11/11 Kara, 2/2 GL, 4/6 ZA
Healing can be variable. DPS can be variable. My job is to be a constant.

n\agrep.exe "gain Holy" Maiden_03.txt
4/29 16:09:41.750 You gain Holy Shield.
4/29 16:09:55.421 You gain Holy Shield.
4/29 16:10:07.515 You gain Holy Shield.
4/29 16:10:35.593 You gain Holy Shield.
4/29 16:10:53.390 You gain Holy Shield.
4/29 16:11:05.250 You gain Holy Shield.
4/29 16:11:17.000 You gain Holy Shield.
4/29 16:11:27.750 You gain Holy Shield.
4/29 16:11:38.984 You gain Holy Shield.
4/29 16:11:59.578 You gain Holy Shield.
4/29 16:12:11.531 You gain Holy Shield.
4/29 16:12:25.250 You gain Holy Shield.
4/29 16:12:36.875 You gain Holy Shield.
4/29 16:12:51.437 You gain Holy Shield.
4/29 16:13:03.062 You gain Holy Shield.
4/29 16:13:15.578 You gain Holy Shield.
4/29 16:13:27.078 You gain Holy Shield.
4/29 16:13:37.625 You gain Holy Shield.
4/29 16:13:50.906 You gain Holy Shield.
4/29 16:14:01.250 You gain Holy Shield.
4/29 16:14:22.937 You gain Holy Shield.
Grand Total: 21 match(es) found.

09:41.8 Delay till next HS
09:55.4 13.7
10:07.5 12.1
10:35.6 28.1
10:53.4 17.8
11:05.3 11.9
11:17.0 11.7
11:27.8 10.8
11:39.0 11.2
11:59.6 20.6
12:11.5 12
12:25.3 13.7
12:36.9 11.6
12:51.4 14.6
13:03.1 11.6
13:15.6 12.5
13:27.1 11.5
13:37.6 10.5
13:50.9 13.3
14:01.3 10.3
14:22.9 21.7

Average delay till HS 14.06

Maiden is an annoying fight for HS coverage due to the silences and the repentance. However, it looks like you're only getting 10/14 or 71% uncrushable coverage maximum. Just one more datum in my crusade to raise awareness in the importance of getting a window for HS refresh.

I have major trouble believing I went 28 seconds between refreshes, even with the silence. Ever since I started logging these and seeing how ugly the crushes are, keeping HS up is priority one.

Still, the logs seem to be pretty comprehensive so maybe I really do suck that badly.

Edit: Yes, I really do. A some silencing trouble initiated the streak, then I caught a Repentance which took a long time to break (this is where she turned around and one-shotted Faroh) then I ran after her and threw JoR which she resisted, then I struggled to get Consecration off (anything to tap her at this point so she'd get off Eruaphadion, the paladin keeping me up), then I pulled her back to the center and FINALLY got HS back up. So a combination of some crummy luck, and me getting flustered because of the crummy luck and fighting suboptimally.

If you don't ban me from the forum, I'll try to improve.
_________________
Scarlet Crusade - 0/49/12 - 11/11 Kara, 2/2 GL, 4/6 ZA
Healing can be variable. DPS can be variable. My job is to be a constant.

Eh - I think warriors often have the same issue, love to see some logs from them.

But if crushes are really a deathknell for tanks, it's important to know that "crush immunity" is really doing it's job. Otherwise we might want to adjust our strategies.

The tankadin simulator could really help with that I think, if we can figure out best case and worst case scenarios for HS usage. That's why I keep bringing these gaps up, I'd love an extra field that says "HS Delay" so I can model the ACTUAL effect of hitting 102.4 (It might be the difference between 85% crush proof and 75% crush avoidance to drop 10% below the cap).