Scattered Shots: Musings on hunter haste design

One of the many goals with hunters in Cataclysm was to make haste a more attractive stat and a more intuitive stat. Back in Wrath, haste was an almost comically poor stat for hunters. Just about the only thing it did for us was increase the rate of fire of our Auto Shots. While it was still a beneficial stat, it was so much worse than all the rest that hunters would bend over backward and jump through hoops to avoid dirtying themselves with it.
Blizzard certainly succeeded in making haste a more attractive stat for us in Cataclysm. It did not, however, score well on making haste more intuitive. In fact, haste is far more complicated and less intuitive than ever before. At certain points, haste is very good for us; then when we get past those haste plateaus, suddenly haste is much, much worse, until we hit another plateau when it's suddenly good again. And of course, these plateaus are radically different for each spec.
While I'm glad we no longer have a nearly useless stat, I do hope that Blizzard continues to tinker with haste to make it work a bit more smoothly and remove some of the unnecessary complexities. In fact, I even have an idea of where to start.
Join me after the cut for Frostheim's haste design theory.
The way designers think
I have a background in the tabletop game industry, and I've had the pleasure of working with some great game designers -- Eric Lang, Darrell Hardy, and Kevin Wilson among them. One of the interesting things about talking with game designers is that they all think about games in similar ways, whether they make board games, CCGs, Facebook games, or computer games.
Game designers are obsessed with simplicity and strategy (well, and fun -- and listening to game designers dissect what makes mechanics fun is fascinating stuff, but another topic).
Designers want to come up with mechanics that are simple and intuitive but have a depth of strategy behind them. One of the very first mantras that they learn way back in game designer kindergarten is that more complicated doesn't mean more strategic, and it certainly doesn't mean better.
Games like chess are a common example of games with incredibly simple rules but amazing depth of strategy. Designers are going for that idea -- they want the strategy and depth, but they want to achieve it as simply and intuitively as possible.
Another interesting thing about working at a gaming company for any length of time is that you get to hear from hundreds of armchair designers -- gamers who think they've made the next great game. The incredibly common theme of all those massive board games (they're always giant, $100 board games) are loads and loads of complicated mechanics upon mechanics, all of which add no strategy or depth to the game but sure as heck make it harder to understand. There's something about the game consumer's mind that equates more complicated with better.
Of course, every time I write a column like this one, I wince a little inside at the thought that I'm becoming one of those armchair designers ...
Intuitive stats
Along with simplicity comes the desire for intuitive mechanics. This gets more important the large and more complex a game is. In a game like WoW, you can't possibly expect a player to read a massive rules tome explaining exactly how every mechanic works. Instead, the overall gist of the mechanics should be obvious just through playing.
Most of our stats are in fact very intuitive. I can see a newer player messing around in WoW for a few hours or a few days just getting them.
- Crit rating? Makes me crit more, right? Right.
- Hit rating? Makes me hit more, or miss less, right? Right.
- Mastery, er (looks at his BM window), makes my pet do more damage, right? Right.
- Haste? Makes me shoot faster, right? Ri ... well, kinda. For some shots, yes, for other shots no, and for other shots, sometimes yes and sometimes no.
Haste does not increase the rate at which you can fire any of your special shots -- though technically, with haste increasing your focus regen, it does in a very non-intuitive way increase the number of focus-dump shots you can fire. Unfortunately, with our base focus regen being so low (at 4 focus per second), haste doesn't have a very large effect on that.
The most noticeable effect of haste is lowering the cast time of our filler shots, Cobra Shot and Steady Shot. Unfortunately, that doesn't actually mean that we fire more of them. Sometimes at certain haste plateaus, we suddenly get to fire an extra one, but a lot of the time we can stack haste and stack haste but still fire the same number of filler shots.
Hunter signature shots
As hunters, our filler shot is gated by our signature shots. These are our super-desirable, big-damage shots that are locked into a cooldown. These shots are great and a fantastic and interesting design, by the way, but because they're so desirable, we rarely want to delay them by more than a very short amount of time.
So if we can fit three filler shots between our signature shot, and then we get 10% haste ... well, we're still firing three filler shots between each signature shot, and then just waiting for a tiny gap of time to fire our signature shot.
Thus, if our signature shot has a 6-second cooldown, then we're firing three filler shots every 6 seconds. But if we can get enough haste to suddenly lower that filler shot cast time to being close to being able to fit four of them in, now we see a sudden leap in shots fired, and our DPS jumps!
Up until that point, the only thing that haste is doing for us is improving our Auto Shot and focus regen (and helping our pet), and isn't a great stat. But when we reach that plateau, suddenly haste is phenomenal for us.
You could argue that this kind of complexity separates the wheat from the chaff and lets the skilled players (or, you know, the ones that read blogs and/or Elitist Jerks) rise to the top. However, more complicated does not make it better. As it stands now, haste is certainly not an intuitive stat, and it adds more complexity than I think is good for the class. I'd rather see players focus on their playing rather than drowning in a morass of specific stat cutoff points.
A streamlined haste solution
Personally, I think the solution here is pretty simple and straightforward -- though certainly it's a dramatic enough change that it'd take a lot of testing and rebalancing of the class:
Allow haste to reduce the cooldown of our signature shots.
Now, along with this, you'd have to set those cooldowns in a way that worked logically with the filler shots (or set the Cobra/Steady cast time in a way that worked logically within the cooldowns). So perhaps your Cobra Shot has a 2-second cast time base, and then your Explosive Shot has a 7-second cooldown base. You can fit exactly three Cobra Shots between each Explosive Shot.
Now you get 10% haste. Guess what? You still fit exactly three Cobras between each Explosive. You are firing both your Cobra Shot and Explosive Shot 10% faster. You get 10% more haste, and you get 10% more shots.
Not exactly, of course; there's our focus dump, which will continue to scale bumpily, and of course our DoTs and whatnot. But now, the majority of our damage is affected intuitively by haste. Your rotation is what your rotation is, regardless of your haste level. It'll still change up based on talent procs, and with more haste, you'll regen focus faster and thus have to fire more focus dump shots.
But now haste would work the way it seems like it should work, and it removes a bunch of gear-juggling complexity that didn't really increase the strategy of the class. Now pair that up with a careful rebalancing of shot damage, and we have a haste solution that I think would be a vast improvement for the majority of hunters.
What do you guys think? Do you like the extra haste complexity that we have now? Or would you rather have a cleaner, simpler solution, as long as haste remained a viable and competitive stat?
Filed under: Hunter, (Hunter) Scattered Shots






Reader Comments (Page 1 of 3)
Kellerune Mar 31st 2011 9:12AM
I personally don't worry much about haste plateaus, and I treat it as my weakest secondary stat. Sure I'm hurting my DPS by not maxing all the way, but there so much more math that goes into reaching a haste plateau, assuming you have certain buffs and such that I'm never 100% sure if going for the haste plateau is right.
If they made haste easier, like reducing CDs of signature shots, haste would have to be rebalanced, damage would have to change, and haste becomes more of a hassle for the developers. I think things are fine the way they are.
Granted, I don't know the exact DPS increase you get from hitting haste plateaus, so maybe that would put it more into perspective.
Boz Mar 31st 2011 10:52AM
I am reminded of an old "Magic: The Gathering" article describing the types of players that designers keep in mind during the course of their work. Designer Mark Rosewater discusses the importance of knowing the MtG audience, and it's worth a read for anyone interested in game design in general:
http://www.wizards.com/Magic/Magazine/Article.aspx?x=mtgcom/daily/mr11b
Kellerune here appears to be a Timmy, a player that goes for power and simplicity. You might not look for the most elegant way to achieve your goals, but you get it done and you do so in a straightforward manner.
By contrast, I like Haste because it appeals to Johnnies. It's complicated, but it's okay to have a complicated stat so long as the benefits from it are incremental rather than game-breaking. You don't have to understand the intricacies of Haste to play a hunter or even to play a hunter well, but when you do the math you will likely see a noticeable improvement.
I don't want needlessly complicated stats and metrics, but I do enjoy the way Haste is currently designed, and hope that its unique qualities persist.
knowlge2 Mar 31st 2011 2:06PM
I completely agree. Its way to mathy and unnecessary imo. Gantis@Stormreaver is my toon and I do everything in my power to avoid stacking haste and with my 4set I still manage to get my CS cast ime down to 1.54 Which is incredibly satisfying on my end and i enjoy it plenty when going marks. Wild quiver procs out the butt!
kingoomieiii Mar 31st 2011 9:15AM
Bah, multiply Haste by ArPen and make it ONLY affect instant shots.
Not, but really, haste needs to be simplified.
tennysonxii Mar 31st 2011 9:17AM
It's an interesting concept, and I agree that it would make haste more intuitive. Of course, you're bound to have those hunters who scream about dumbing down the game up in arms. I can't really think of any downsides, but I'm sure your beloved commenters will be quick to point them out :P
mibu.work1 Mar 31st 2011 9:19AM
I'd simply like to be able to look at haste and explain why my spec (BM) seems to provide so much of it to my MM friends, and why this is a good thing, so yes, I'd like a simpler haste system with more intuitive use. Personally, I just consider haste to be on level with mastery and crit when I'm looking to reforge some hit in, and that's all I consider it for. I know that it theoretically makes me more powerful, and it sure as hell is awesome when I pop rapid fire, berserking, bestial wrath, my item CD's, and bloodlust is up. At that point, my cobra shots are at about a .2sec cast time, and my auto-shots never seem to stop. Heh, it's practically eight seconds of +4k DPS for me, if I don't have to move. Moments like that put me as our guild's #1 damage dealer.
CrossEyed Mar 31st 2011 10:07AM
+4k DPS puts you at the top for your guild DPS??? Your guild sucks :-P
jasonjcohen Mar 31st 2011 10:55AM
Bloodlust + RF is a no-no. your .2 second cobra shots mean you spend roughly .8 seconds of your GCD not doing anything. Separate those and you'll see a further DPS jump. and I think he meant and *extra* 4k of DPS over his base.
mibu.work1 Mar 31st 2011 11:17PM
Okay, how do you reccomend doing that Jason?
And yes, I'm usually in the 11-12k DPS range on boss fights like Regular Maloriak or Omnitron (discounting Green phase and the power wells), but if I time my Cooldowns to coincide with bloodlust, I can reach the 15-16k range for a few seconds, assuming I enter with 40+ focus and kill command off cooldown. The variance is due to the possibility of Focused Fire and Killing Spree proccing, as well as normal crit chance when using a devilsaur pet (+5% crit chance).
Admittedly, my DPS is not stellar because I'm somewhat lowly geared, but I tend to be better at surviving than my fellow raiders, hence why I'm one of the better DPSers: You put out more DPS if you're alive for the whole fight.
duskhawk Mar 31st 2011 9:19AM
I don't see haste going that way because rotation macros would be mindlessly easy then. I think Blizzard would want to avoid the kind of mindless macro spamming that happened with BM in TBC. Not that you can't do a cast sequence now, but it's not as easy as knowing it will always scale perfectly with haste.
Frostheim Mar 31st 2011 10:21AM
No, you still won't be able to use rotation macros -- every spec has different abilities that interrupt their rotation (Lock&Load, Master Marksman, Bestial Wrath) and even without those, we still need to alter our rotation to fire more or less Arcane Shots depending on the state of our focus.
Keep in mind that right now haste doesn't make our rotation fluctuate mid-fight (except for rapid fire/ancient hysteria) -- it just changes based on our gear and what buffs we have, but then we're using the same rotation all fight long.
Mark Mar 31st 2011 9:52AM
I love how cata simplified all classes mechanics, and i love the current hunter balance, but changin the haste mechanic will just be sorta screwup thats gonna be fixed and fixed and fixed all over again. Its not tough to understand right now and the concept of haste plateus and their mechanics, its imo perfect, it keeps people thinking, lurning and keeping themselves up to date with their class, i wouldnt wanna see excessive dumbing down amongst Warcraft universes best class :)
Loronar Mar 31st 2011 9:28AM
If haste only affected hunters, that could be an interesting solution. However, because haste affects everyone, I can see why it's such a complex statistic. For example, compare hunters and a caster class, say mages. Hunter auto shots now function on a layer different from our special shots. On the other hand, there is no such thing as a mage auto attacking with his/her wand that is on a secondary level to the mage's special attacks. From my point of view, it's the current mechanic of auto shots that has made haste such a complicated statistic for hunters because single-lineup shooting no longer exists.
Zarst Apr 1st 2011 3:57PM
I like this idea. but I wonder if it would eventually end up with me having to keep up with my rotation by hitting my key binds much faster as haste levels increase. I feel like it could increase the stress of keeping the rotation together, as well as tire my fingers, which doesn't seem very fun (then again, maybe it would be). Can't say I can come up with any better ideas though, and I do wish haste mattered more, so I'd be willing to give it a try ^_^
styopa Mar 31st 2011 9:49AM
Er, why not just have haste increase the rate at which we gain focus more meaningfully?
It's MUCH more intuitive, it is plateau-less, and utterly ignores other stats, gear, etc.
In re your comments about game design, I think Blizzard has amply and repeatedly proven that while they know a lot about designing overall great games, their implementation of mechanics has always been clumsy, Byzantine, counterintuitive, and, well...broken. I don't expect that to radically change anytime soon.
From the great "ratingization" (where they took relatively simple "+1% to crit" and changed that to a "rating" that would then scale negatively against a player's increasing level), to the giant +STA payout at the start of TBC (and now again in Cata), ArPen, haste, etc. it's happened over and over. Blizz can be totally exonerated from blame in vanilla - nobody could have predicted that WoW would be as huge and persistent as it's been. But at that point, to rewrite the mechanics and fail to future-proof them? And then again in new mechanics introduced in WotLK? And now again in Cata?
There are dozens, if not hundreds of nicely scaling combat-system mechanics developed for RPGs for the last 30+ years that they could have modeled their system after. Instead, they chose to build their own table-based system that was prone to breaking as ever-spiralling buffs pushed results off the table....an intrinsic problem of table-based systems forever.
TL:DR - Blizz has never been good at future-proofing mechanics, I don't expect that any change they implement now will be different.
tennysonxii Mar 31st 2011 11:47AM
Maybe it's just because I play solely as BM, but the only time I ever have to worry about focus is when I'm about to cap. In that case, an increase to focus regen would simply mean more focus dumps, and probably wouldn't be a huge benefit.
Zippö Mar 31st 2011 9:55AM
to be true, i don't think it's too complicated, though.
in this set amount of time Y, i can fire of X fillershots, and i will only be able to fire off X+1 shots if i get Z haste, or more clearly if my fillershot has a cast time that's smaller than Y/X, so to speak if it is Y/(X+1).
so if my math isn't completly off, this would mean for sv, in 6 seconds i can fire off 3 filler shots with a cast time of 2 seconds, and can only make it 4 shots when they have a cast time of 6/4=1.5, so i would need 25% haste to cut of this 0.5sec cast time.
correct me if i'm wrong, but to me it's quite logic and easy.
that's the difference between the casual and the nerd, or the huntard and the awesomesauce of hunters. we don't walk into a raid with a faceroll tactic, we go in there with knowledge and an understanding of our class. i think this should be rewarded.
and keep up the great work frost :)
Frostheim Mar 31st 2011 10:21AM
Not exactly, no.
First of all, your instant shot triggers the global cooldown, so you only have 5 seconds of active time, not 6.
Secondly, you get haste effects from various sources -- haste on gear, raid buff, talents, etc. Those don't stack additively -- so if you have 10% from gear and 10% from a raid buff, that's not the same as having 20% haste.
And thirdly, here's the complicated part: if you can fit 3 fillers between signatures, your next plateau *isn't* when you can fit in 4 between -- it's actually the point that you fire four, but push your signature back *just* enough that you're still getting a net gain (this is usually as much as a second of delay) -- but that point is different for everyone depending on your gear. Calculating that plateau is pretty annoying, and not something that vast majority of players are ever going to do. And it's certainly complicated as all heck : (
Zippö Mar 31st 2011 10:46AM
as i see it, that's when you make it complicated.
i read your article on haste plateaus over at the WHU several times, and came to the conclusion that, given i don't have the 4 set bonus, and would never reach the soft plateau through reforging, it's better for me to reforge into crit.
i would suppose this holds true for most people.
so while there actually would be this squishy soft plateau which would have me gain a bit more dps through haste, i would ignore it until i get the 0.2 sec reduced cast time which brings me much closer to the next plateau. and from that point each point of haste actually gives me a dps gain, while i've already reached the soft plateau with 4 set bonus+current haste.
the annoying thing with haste is how it shifts in the stat priority from desirable, to useless, to desirable again.
as response to your point 3) Frost, i was referring to the hard plateaus obviously, the equation would without a doubt be a bit more complicated for the soft plateau :)
Boobah Mar 31st 2011 2:13PM
"so if my math isn't completly off, this would mean for sv, in 6 seconds i can fire off 3 filler shots with a cast time of 2 seconds, and can only make it 4 shots when they have a cast time of 6/4=1.5, so i would need 25% haste to cut of this 0.5sec cast time."
Which, I'm afraid, it is. If you can get four shots off with haste in the time it takes you to get three shots off with no haste, you're getting (4/3) = 133% as many attacks, which is +33% haste. 25% haste only gets you to (2/1.25) = 1.6 seconds on your non-hasted 2-second cast.
Which leaves aside the previously mentioned second of downtime after each instant attack. And the also previously mentioned fact that sometimes it's worth pushing back your signature shot a little. As Frost said, it's complicated.