More on Black Arrow and Lock and Load
by Mike Schramm 
Jun 9th 2009 at 2:00PM
The Hunter's Mark examined what's going on behind the Hunter talent Lock and Load a little while ago, and last week Atkallen did a follow up post getting down into the nitty gritty of the numbers a little further. Essentially, he charges that you're looking at an LnL proc for about 2-4% uptime during a four hour raid -- that is, for about 4% of the time, you've got Lock and Load giving you free Explosive Shots with no cooldown and no mana cost. And as he says, that's the "applied" time -- since LnL disappears as you use it, odds are that the real time you have it available will be much less.He also tracks that time from 3.1 to 3.1.2, and says he's found data that confirms his previous theory: that it was more than the tooltip that needed changing during the last patch. Before the patch, he was looking at 11% uptime during three Ulduar fights, and after the patch, that's dropped down to 6%. Blizzard said they were only changing the tooltip, but that doesn't seem to be the case.
Finally, he stands by a notion that a lot of our commenters debated last time: that LnL is proccing more after 3.1, even though the mechanics claim that it shouldn't be. I haven't had a lot of personal playtime with it post-patch, but there are two different conclusions to take away here. First: LnL, in conjunction with Black Arrow and Resourcefulness, is a definite force for DPS in the Survival spec no matter how often it's proccing. And second: there still may be something else going on here behind the scenes -- the removal of the cooldown is almost definitely a factor, and Blizzard may be playing around with the RNG more than they're letting on.
Filed under: Hunter, Analysis / Opinion, Odds and ends, Classes, Talents
Reader Comments (Page 1 of 2)
Ahima Jun 9th 2009 2:15PM
I hear Onyxia is deep breathing more :p
Reinard Jun 9th 2009 2:47PM
This.
Bubsa Jun 9th 2009 4:04PM
Hahaha
Classic!
TEJ Jun 9th 2009 4:09PM
And that.
Rishio Jun 10th 2009 1:56PM
I honestly don't know much about lock and load pre 3.1 I respecced after for raiding.
Zebulin Jun 9th 2009 2:25PM
This is what I've been saying for the past month or so. I even have numbers to back it up, as I included the last time this was brought up.
LNL has around a 4% uptime over a 4 hour raid. It used to be much higher.
Onyxia really is deep breathing more.
Nick S Jun 9th 2009 3:02PM
Once upon a time, Lock and Load reliably procced 2-3 times for me on a long fight.
Now, I'll have a fight where I see 5 procs, 3 of which comes one right after the other. My DPS skyrockets. Then the next fight yields zero procs, and my DPS is pedestrian. It's a very noticeable difference and a very weird mechanic now.
QQinsider Jun 9th 2009 2:38PM
While there seems to be a lot of anecdotal evidence flying around that something has changed, there is no proper data presented here. Small sample sets on single boss fights (or even over a raid, where other factors like downtime are so variable) and comments like "at the Heroic Training Dummy and my Lock and Load procced twice in a short amount of time" aren't very convincing.
outdps Jun 9th 2009 2:57PM
It's not the anecdotal characteristic I'd worry about, it's that we're measuring the uptime on a buff that, when used effectively, disappears in 6 seconds. For bad players, that's more like 4.5 seconds, and at the end of a fight, it can run the full length.
Nick S Jun 9th 2009 3:05PM
I've gotten strings of three and four LnL procs, not close to each other, but literally continuous uptime during a boss fight, using no special tricks. That's anecdotal, but it nonetheless demonstrates that its possible to proc an insane number of LnLs in a row, which implies that it's merely a random chance with no ICD.
Nick S Jun 9th 2009 3:07PM
...continuing since the 2nd half of my comment was eaten.
Given that I've not changed my playstyle in any significant way, the fact that I'm seeing fewer total LnL procs on longer fights seems to indicate that the proc chance has changed. I haven't done a statistical analysis, but I can look at my WWS reports from raid after raid and see intuitively that the average number has dropped.
outdps Jun 9th 2009 3:26PM
@Nick they removed the cooldown- I've even gotten procs while the first proc hadn't been used up yet. That alone increases the average number of procs we can get per minute of black arrow uptime. The key is data though. I'm going to see if I can find a way to parse combat logs to record the proc chance per tick of the BA dot. Shouldn't be to difficult.
outdps Jun 9th 2009 2:55PM
The real question is whether the proc chance per tick of black arrow has changed. That metric is the only thing that blizzard controls- the rest of it is up to the player. I'm working on collecting data for that, so if anyone has anything from pre or post patch they'd like to share, drop me a line: http://outdps.com/contact/
Heilig Jun 9th 2009 3:04PM
This.
Uptime means nothing. if you don't keep black arrow up at every cooldown, your numbers are off. If you drop a single trap, your numbers are off. Uptime means nothing. The only thing matters is the NUMBER of procs compared to the number of Black Arrow ticks, which seems an easy thing to determine from any WWS report.
swampsquatch Jun 9th 2009 3:19PM
So this is why my BM hunter out dps's better geared SV hunters. If you get REALLY unlucky and don't proc LnL, dps goes:
|
|
|
|
\/
I find the mechanics really interesting though, I wish someone would figure out what the deal is with it while I farm gold for dual specs.
outdps Jun 9th 2009 3:24PM
Actually that's likely a shot priority problem for the SV hunter. I've seen 10 minute Ignis fights without a single lock and load and my DPS suffered, but not nearly as much as I thought it would. To be honest, it's still a good talent, but it's no longer the bread and butter of the SV tree.
TEJ Jun 9th 2009 3:56PM
Christ, can you please stop linking blog posts from bad hunters with no idea in statistics?
Seriously, everything written there is FUBAR.
Knob Jun 9th 2009 4:15PM
Exactly. I thought this new editor knew what to write about hunters going by his previous article, but he's again done the same thing as his predecessor; link stuff which are baseless and anecdotal combined with only a miniscule amount of parses being passed off as empirical data.
Personally, I went MM a bit before the 3.1.3 patch and now am full-time MM so the clusterfuck that is the LnL proc rate doesn't affect me anymore. I can't say that's a bad thing too.
TEJ Jun 9th 2009 4:36PM
^ and this. SV is too RNG-based: EW, MT, LnL. Of course, they can be averaged out but when LnL procs are the difference between #1 and #2 on recount, it is indeed serious.
On that note, am I being too pessimistic by declaring we have a new BRK to create even more unrepentantly bad hunters?
Kheldul Jun 9th 2009 4:03PM
LnL uptime is orthogonal to the whole proc frequency issue. LnL "uptime" could be maxed if you always drop traps (never BA) and never actually use the LnL before it expires. I attempt to use LnL as efficiently as possible. I may proc it from BA or traps.