The particular context here is the idea that you are faced with two choices:

* One that directly increases my damage (e.g., AP and/or crit)
* One that increases my hit%

With just that much information, THERE IS NO RIGHT ANSWER. You need to know how much of an increase to your damage vs how much of an increase to your hit%. Even if you had that information, comparing the two choices is a decidedly non-trivial task that has to factor in a very large number of variables. Which is why a lot of us are saying not to try to use a simple rule of thumb like "Get hit capped before worrying about anything else". Instead, the right answer is "Plug both choices independently into the spreadsheet and pick the one that gives you the higher resulting dps".

Say what? Any sensible reason you don't use the Crit scope?

Bah, i have made my arguments, if I can't convince someone to always go for the hitcap so be it (with always i mean a high end raiding hunter in full Naxx/Ulduar gear, not a lolhunter who has just dinged 80 and wants to enter Naxx)

Like someone else said on this page; you either choose to go for higher dps by following the spreadsheet (which is still 'just' a mathematical approach to the real deal, you cant make programs to predict the weather on the long term either), or choose to go for better dps (read: consistent dps, knowing you wont miss any shot...ever)

If your guild/raidleader asks you when he brings you to Yogg Saron hardmode 'can you do good dps?' I would rather say I do a consistent amount of dps, opposed to saying I 'can' do alot of dps, but not guaranteed. (Ofcourse when your guild reaches YS that leader should have a pretty accurate idea of what your dps is like )

On a side note: you said the spreadsheet, if correctly modeled, accounts for those 'reality' situations i described earlier. Apart from the fact that chance is very hard to model, the model will look at the differences between 98% 99% and 100% as just numbers. It doesnt recognize the 100% hitcap as a magic number where suddenly no misses will occur. Perhaps people with more understanding than me of how the spreadsheet actually calculates dps can reply here.

One more example, because i like examples: we have three hunters, hunter A with 98% hit, hunter B with 99% hit, and hunter C with 100%. They each have similar gear (same iLvL, not BiS, since 2 of them arent hitcapped), do around same dps each week (which is perfectly possible ofcourse, even though two arent hitcapped), and all 3 have the Darkmoon trinket. Only hunter C knows that his trinket will proc the first second he attacks the boss, only hunter C knows the trinket will proc again 45 seconds later, and again 90 seconds into the fight. Hunter A and B, both not hit capped, have no guarantee that they will hit.

This is for me the most important reason to be hit capped; procs depend on hits, not on damage done, and the spreadsheet 'only' calculates the average uptime. The trinket doesnt say you need to do 20000 damage in 5 seconds to proc it. No, it says you need to hit. When you miss a few shots 45 seconds into the fight (and believe me, you will at some times) you will not refresh the trinket proc (I could have taken Expose Weakness as an example too ofcourse). This in turn also means that the shot you fire at 48 seconds finally procs the trinket, but this shot wasnt affected by the DM trinket buff in the first place. Hunter C however knows that the shots he fired 48 seconds into the fight, were in fact affected by the DM trinket. You delay all your procs by not being hit capped. That's what i meant earlier by saying that any possible upgrade you could have chosen over the hitcap, would have been present in the very hitcap itself, in the form of maximum possible uptime of procs.

I don't like RNG, any opportunity to eliminate it is a chance worth taking. Anyway, i believe we can both see that at the end of the day the spreadsheet will usually tell you to be hitcapped in order to do maximum dps (this goes for endgame gear); it's why you and i are both hit capped. What started this argument was a question from a hunter asking whether he should go for the last 2% hit he needed, or take another gearchoice giving 4-6% damage increase. I would say the first no matter what, you probably the last.

Some procs activate when the attack is made instead of when the attack actually connects with the target.

Given a correct model, 2 different gear sets with the same DPS listed (using e.g. one gear set favoring crits and procs another favoring Hit Rating to cap and Attack Power), both will result in the same DPS if averaged out over a long enough period of time.
In such cases, it's down to personal preference. Some prefer consistency over high potential, and vice versa. There's not really a point arguing for several pages about what's the "right" choice.

Using the term "better DPS" while arguing for Hit is also misleading. Better in the case of DPS meaning higher, consistent meaning with less variation.

The examples with missing several Chimera Shots in a row forcing you to use a global cooldown on Serpent Sting again, or other adverse affects of missing are valid, but those concerns regarding the validity of the model used, not magical reasons for why the "right" thing to do is to always and without question cap Hit.

@ Lactose

I disagree that better dps necessarily means higher dps.

When I use the term better dps, I mean better - and better can be different from higher when you can deliver the dps "on command".

I think the term should be "consistent" DPS than, not "better". Plus, I think a lot of the people making an argument for "consistent equals better" DPS are really exaggerating a bit here. Having 99% hit is not going to throw off your DPS, even under relatively bad RNG for a fight, by that extreme of an amount. I don't mean to put words in anyone's mouth, but some of the people are making it seem that having 99% hit means your DPS is going to swing wildly from fight to fight, and that's simply not true. Yes, there is the possibility that the RNG gods could in fact despise you, and you could miss a whole lot one fight. I suppose it's possible. But a situation like that is hardly realistic, and it begs the question of how you feel about Crit as a stat too, as it's really not any different in terms of RNG than hit.

A lot of this argument is a moot point, though, simply because before you're hit capped, gemming for hit is 99% of the time a DPS boost according to the spreadsheet. Most hunters with anything close to decent gear should be able to get the hit cap through enchants/gemming at the very least, and the spreadsheet will almost always consider it a DPS upgrade There are exceptions, of course, most notably when people are VERY close to the hit cap and a hit gem would put them over, which is the situation I'm in. The spreadsheet says that a +16 agi gem and staying at 262 hit is better than a +8 hit/+8 crit gem and going over to 270 hit, albeit the difference is very minimal. Since the difference was so minimal for me, I chose the "consistent DPS" route (note that it's not the "better DPS" choice, simply slightly more consistent.)

Not to ramble on either, but I've heard at least one person say the spreadsheet can't tell that 100% is a "magic number" for hit. That's because, in terms of DPS, it's not magical at all. Yes, it takes a slight RNG out of play, but there's still so much RNG left (critical hits being the main thing) that you're hardly taking luck totally out of the equation. There is nothing magical from a DPS standpoint about 100% hit, although you may magically feel better about reaching it.

If pets would ever get our Hit rating and not our percentage, then Hit will take a nosedive when we reach the point where a gem would send us over the cap. Wasted stats are wasted stats.

RNG can work both ways, and I consistently ran with less than the cap in TBC. Not because it was a choice always, but even 15 ratings below the cap then I had many fights where I didn't miss even once, and most not more than once or twice. So arguing about the insane 'several Chimeras' is like arguing that Greatness won't proc for minutes because it only has a 35% chance to proc. That is to me just nuts.

The RNG works like this at this level (we are talking people who are just not capped, not people who haven't used Hit at all here), mostly the not-capped Hunter will do more damage, but there is a relatively small chance he/she will do less. In either case it is a very small difference from a fight perspective. And a missed Chimera will certainly hurt a lot on the rare occation it does happen.
So yes, I agree that now in Wrath Hit Rating is slightly more important for the reasons that we have a lot of damage invested in specific single attacks, and should those miss we would see a fair bit of DPS loss (on that given fight). Meaning the potential loss of missing can be unreasonably high compared to the overall gains of harder hits.

Sorry, but you are (boldly) incorrect here. The spreadsheet can absolutely say that a gear choice which leaves you below the hit cap will provide higher average dps than an alternate choice that leaves you above the cap.

It sounds like you are somehow taking issue with the notion of average dps being the proper milestone. Instead, you want to use guaranteed dps. But as I (and others) have pointed out, such a metric does not exist. There are too many other random factors at play -- the largest of which is probably crit rating, but partial resists and proc rates on trinkets and abilities like LnL and EW all have random components. It simply makes no sense to choose one of the many, many sources of randomness and treat it as some sort of magical "special case".

As others have pointed out, this argument is unlikely to lead to much of a practical difference in gearing. Gear choices generally come incrementally, and given that hit rating (for hunters!) is generally valued higher than other stats in our mathematical models, we'll both generally opt for gear that has more hit (until capped). However, there is a larger principal at play here. This is fundamentally a theory-crafting site. We (as a community) place a high value on understanding the technical mechanics at play in the game and in building/using relatively accurate mathematical models to optimize our game play. Making an argument that you should selectively ignore the models when you don't like what they tell you is the moral equivalent of saying "crit is better than AP because your shots hit harder".

OTOH, maybe you actually have a valid objectively quantifiable argument for how hit is mistreated in the spreadsheet. If so, make your case, and let's get it fixed.

To me, this does make sense, since hit is one of the few RNG elements you can in fact eliminate. And yes, you are right, it is just one of the few RNG elements you can get rid off. No hunter will ever reach 100% crit apart from Loatheb encounter. And yes, any hunter with 2/3 Expose Weakness instead of 3/3 will not see a 100% uptime of EW, but a 95% uptime on average. Same goes for all our other procs and buffs/debuffs. But like i said before, i highly doubt the spreadsheet will ever tell a Naxx/Ulduar geared hunter to go 2% under the hitcap in favor of a similar item that gives a similar amount of agility (i.e. 60 agility opposed to 60 hit rating).

The only reason we are even discussing this hypothetical case is because one hunter (I'll edit his name here in a moment. Edit: it was Daragh, almost at top of page 11) was asking what stats he should go for ; 2% hit to reach the cap, or more agility/crit/AP instead if the spreadsheet claimed the latter would give more dps. And i am saying that the spreadsheet will most likely say that going for that 2% extra hit to reach the hitcap would be the bigger dps increase, since hit will give a relative dps increase no matter what your gear is.

A hunter with 99% hit and doing 4000 dps will go to 4044 dps with that 1% extra hit. Another hunter with 99% hit and doing 7000 dps (Ulduar gear) will go to 7077 dps with that extra 1% hit. If the same two hunters had gone for agility instead, the first hunter would have seen a bigger relative dps increase. So in a way you are right Alarge, if the spreadsheet tells you this agility upgrade is a bigger dps increase on average, you should probably go for it. However, this does mean your gear hasn't reached a certain level yet (pre-Naxx i would say, since my spreadsheet insisted on going for the hitcap, and i believe so did yours), and you still havent eliminated one of the many RNG elements

It might make sense to you (to eliminate the variable), but that is based on emotion, not mathematics.

Let me re-iterate: I'm not arguing against hit. I'm arguing against prioritizing hit higher than the spreadsheet does. The spreadsheet will almost always (depending on your gear) prioritize 60 hit over 60 agility if you are 60+ points under the cap. The 60/60 point is not relevant to this discussion. Nor does it make any sense to talk about gear choices you make at 4000 dps persisting until you are hitting 7000 dps. Each time you get a new piece of gear, you need to re-optimize your entire gear-set as a whole -- this might mean re-gemming, re-enchanting, switching in earlier discarded pieces, etc.

But suppose the choice was between 60 agility and 30 hit? Or 40 agility and 30 hit? My point is that there is no simple "rule of thumb" you can follow and end up with the right answer (e.g., always go for hit until you cap) -- the interplay between the variables is too complex for a simple rule. The right answer is "plug both choices into the spreadsheet and pick the choice with the better resulting dps".

Pitching hit capping as eliminating a source of unpredictability is intellectually dishonest. Instead, you are advocating lowering your average dps in order to provide marginally more consistent dps. Put like that, the argument is less attractive, isn't it? (but more intellectually honest).

FWIW, I decided to play with the stats a little and see what 1% difference in hit would mean for me. Basically a loss of 33 hit (~1%) would result in loss of ~0.78% of my total dps. A loss of 33 agility would result in a loss of ~0.74% of my total dps. So 1% hit isn't a simple conversion to 1% of dps. And, while hit is more valuable than agility, the crossover is at about 1.06 hit per point of agility. So, for example, 32 agi > 30 hit for me. I contend that most of the "get hit capped first" guys would take 30 hit over 32 agility. And they'd be wrong (zing! )

Yes I would take 30 hit over 32 agiif I was 23 or more hit rating from the cap - and no, I would be right.

This statement (Iknow you meant it as a zing) is blatantly false, as the value of agility (ap and crit) is relatively lower if you are missing 30 hit, and thus the 30 hit is worth more than than the extra 2 points of agi.
It is also false because agi only provides 1 ap per point, and I am BM specced, so I gain more from ap than I do from agi per budgetpoint, and 32 agi is not attractive since its only 32 ap, and half the crit value as crit doesn't transfer to pets.

So that statement was wrong in my opinion because I advocate consistent performance over averagely better performance - AND because it was not based on actual data.

Try and lower your hit by 30 under cap and see what brings your dps up more on average (the spreadsheet calculation way) - I am pretty sure that you will not only get consistently better dps, but averagely higher dps from taking the 30 hit rating assuming you use 30 hit rating of the 30 hit rating