Attack Rating isn't rewarding enough

One of the reasons that dex isn't valued highly relative to str or int is that it simply isn't very effective on any level. Evasion% obviously comes up short relative to life% or ES%, but even worse is that the returns on accuracy.

I have a level 35 ranger with 240 dex and over 150 accuracy on gear, and I'm only hitting 93% of the time. That's ridiculously bad for the amount of investment into accuracy rating the character has made. It's not even 95%, which means I'm missing more than 1 time in 20. And this character is seriously over-invested in dexterity and accuracy for her level.

A single passive skill point can buy people on the other end of the tree Resolute Technique, and they hit 100% of the time. Sure, they give up critical hits, but a character who is investing heavily in dexterity doesn't have very impressive critical hits anyway. (Indeed, the only builds that can reasonably pile on critical hits are spellcasters, because they don't need to worry about to-hit rolls!)

Accuracy investment needs to be able to outperform Resolute Technique, or its just obviously inferior to resolute technique in every way. Because to come even on expected damage with 95% accuracy requires +50% critical damage or 10% crit rate, and that's a substantial investment in skills even beyond the dex/accuracy investment.

Thus
(1) To hit % from accuracy needs to be able to reach 100%, and not in some unreachable limit condition
(2) Extra accuracy beyond 100% should translate into additional critical chance or critical damage at some rate, representing superior critical performance from high accuracy.

"

Squirrelloid wrote:

(1) To hit % from accuracy needs to be able to reach 100%, and not in some unreachable limit condition
(2) Extra accuracy beyond 100% should translate into additional critical chance or critical damage at some rate, representing superior critical performance from high accuracy.

These are good ideas - although any boost to accuracy is a nerf to evasion, so evasion would have to be buffed too - a never ending cycle! Unless GGG were happy to use different mechanics in pvm vs pvp; but I don't think that's how they want to build the game.

I guess they could make it more linear, so basically it would be a boost to both high evasion and accuracy, and a nerf to low evasion/accuracy.

Also you are totally underestimating crits. Crit attack builds are definitely viable, even without Lioneye's Glare. Especially now that they buffed Critical Weakness.

"

dudiobugtron wrote:

"

Squirrelloid wrote:

(1) To hit % from accuracy needs to be able to reach 100%, and not in some unreachable limit condition
(2) Extra accuracy beyond 100% should translate into additional critical chance or critical damage at some rate, representing superior critical performance from high accuracy.

These are good ideas - although any boost to accuracy is a nerf to evasion, so evasion would have to be buffed too - a never ending cycle! Unless GGG were happy to use different mechanics in pvm vs pvp; but I don't think that's how they want to build the game.

I guess they could make it more linear, so basically it would be a boost to both high evasion and accuracy, and a nerf to low evasion/accuracy.

Also you are totally underestimating crits. Crit attack builds are definitely viable, even without Lioneye's Glare. Especially now that they buffed Critical Weakness.

I was thinking more linear. Doesn't need to particularly be a nerf to low accuracy, although it does need to be a nerf to low evasion. (Whether or not its a nerf to low accuracy depends primarily on where the intersect is between the current and new accuracy equation).

I don't know what the current equation is offhand, but you could probably get pretty close to the current equation over most of the 'interesting' part of the range, and the major deviation in output would be at currently low and high chance to hit. (I'm guessing there are long tails at either end where changes in accuracy don't result in much change in chance to hit, so mostly you end up making those tails more linear).

The other thing you can do is just take the current equation shape and extend it across 100% to hit, so the serious diminishing returns at teh top end happen after 100% to hit is achieved. (This would instead limit whatever benefits you get over 100%).

I agree with you that accuracy is pretty difficult to get in sufficient amounts. The passives on the tree are terrible when it comes to providing additional accuracy. Normally when I take a 20% accuracy passive, the chance to hit goes up by 1%. And then it goes back down again next level, which really feels like a total waste.

I think a lot of it has to do with the scaling of accuracy. It seems like it has some pretty ridiculous diminishing returns. Frenzy on my level 61 ranger has an 86% chance to hit with 334 dexterity and 1362 attack rating. If I add my Additional Accuracy support to it, the attack rating goes up to 2381 and the attack has a 92% chance to hit. So that's about a 75% increase in attack rating for that 6% greater chance to hit. So it's no surprise that the 8-20% accuracy passives on the tree aren't very effective.

I don't understand the 5% minimum evasion chance or the 95% maximum hit chance either. Seems arbitrary and unnecessary.

"

(Indeed, the only builds that can reasonably pile on critical hits are spellcasters, because they don't need to worry about to-hit rolls!)

Certain non-spellcaster builds or items can remove the reliance on hit rolls too. Rain of Arrows always hits. Attacks with Lioneye's Glare and Reaper's Pursuit always hit. Projectile Weakness debuffs monster evasion against projectiles, resulting in almost no evading.

Strength-based melee characters are in the most disadvantageous spot in regards to accuracy in many ways. They don't have a lot of dexterity, which means that they don't have a lot of base accuracy. The example above demonstrating how worthless accuracy nodes was on a dexterity character with good base accuracy, so you can imagine how completely worthless the percentage-based accuracy nodes in the strength and intelligence areas are going to be with hardly any base accuracy to improve. Melees don't have a curse like Projectile Weakness to increase hit chance either.

I'd really like to see them go around and change all percentage-based accuracy nodes in the strength and intelligence areas to give base accuracy, and then maybe add in an intelligence/strength aura that gives a percentage increased accuracy. The percentage accuracy nodes in the dexterity areas should remain percentage-based, but should really be buffed.

Wouldn't it be simpler just to reduce the evasion rating of monsters, thus giving accuracy a relatively stronger effect?

However, I'm not really sure that the problem is accuracy being too weak. There are just too many ways to bypass it, and the things that bypass it are not any weaker than the things that have to go through the accuracy mechanics (often they are stronger outright). I see it as two separate problems:
1) Weapon attacks that have to deal with accuracy should be noticeably stronger than spell attacks that bypass accuracy, to compensate for the "accuracy tax."
2) Resolute Technique is too powerful for the number of points it costs to take the node. The reason this is problematic is because it prevents Issue 1 from being addressed. Using this node, weapon attacks are (almost) just as easily able to bypass the "accuracy tax."

"

dudiobugtron wrote:

"

Squirrelloid wrote:

(1) To hit % from accuracy needs to be able to reach 100%, and not in some unreachable limit condition
(2) Extra accuracy beyond 100% should translate into additional critical chance or critical damage at some rate, representing superior critical performance from high accuracy.

These are good ideas - although any boost to accuracy is a nerf to evasion, so evasion would have to be buffed too - a never ending cycle! Unless GGG were happy to use different mechanics in pvm vs pvp; but I don't think that's how they want to build the game.

I guess they could make it more linear, so basically it would be a boost to both high evasion and accuracy, and a nerf to low evasion/accuracy.

Also you are totally underestimating crits. Crit attack builds are definitely viable, even without Lioneye's Glare. Especially now that they buffed Critical Weakness.

Only in PvP. I would suspect that monsters do not have Strength, Intelligence, or Dexterity scores but instead have their derived attributes defined directly.

"

PolarisOrbit wrote:

Wouldn't it be simpler just to reduce the evasion rating of monsters, thus giving accuracy a relatively stronger effect?

That totally depends on the shape of the accuracy/evasion curve. If its impossible to breach 95% to hit with the current math, then making accuracy stronger but otherwise using the same equation *cannot* solve the problem. 100% to hit has to be achievable with accuracy.

If 100% is achievable without the keystone or LEG, both of those things become obsolete. Either get one of them or deal with the fact you cant hit every time.

Resolute technique is fine as it is. Giving up crit for 100% hit chance is a steep enough penalty. There's hardly any builds that are OP because of resolute technique, especially when you consider that spells have 100% hit chance anyway and they can stack crit to pretty insane levels.

That said, accuracy needs a bit of a buff. Getting to 95% hit chance with accuracy needs to be feasible, instead of the current system where you can stack accuracy on every piece of gear that can get accuracy, invest 5-10 accuracy nodes, and still only top out at about 85-90% accuracy at higher levels.

Buffing Attack Rating doesn't work, because it makes the Evasion weaker, and it's already weak enough... either that, or it makes it too strong.

Downright paradoxical how that works. But it's the truth: take some of your "attack rating over 100%" situations and try attacking them from an evasion standpoint, pretend to be the defender. Linear solutions simply don't work.