The big melee DPS issue: Grazing Hits

[Cetus]

Question: Do you think a nearly maxed to-hit fighter should have a 75% success rate in landing their melee attacks on end-game bosses? Here's the way things are currently working for me:

GRAZING HITS

Grazing hit formula as it stands:
(To-hit bonus + 10.5) / (2 * Target AC) + 25% for proficiency = % chance to hit rounded to the nearest 5% (this is determined by the die roll you see in your combat log).

R10 Black and Blue end dragon boss with a full improved destruction (-15 to AC):

Rolled a 5 with a 176 to-hit: Graze



Rolled a 6 with a 176 to-hit: Hit



This means on rolls 1-5, I will graze with a 176 to-hit. This means that I have a 75% success rate in landing my attacks on this target.

This means that the dragon has about a 190 AC, after incorporating a -15 to it's AC with improved destruction (since I was using that at the time of this calculation).

Just to check my math: 176 to-hit + 10.5 = 186.6/(190AC *2) + 25% = 74% rounded to the nearest 5% = 75%

With this formula, how much additional to-hit or target AC reduction do I need to not graze on a 5?

Landing a hit on a 5 means I need to have an 80% success rate, with a 190 target AC I need:

(10.5 + x To-Hit) / 380 + 25% = 80%

x = 380(.8 - .25) - 10.5

x = 198.5

I need to increase my to-hit by 20 additional points (currently impossible) to land attacks 80% of the time.

What about an 85% chance success rate?

x = 380(.85 - .25) - 10.5

x = 217.5

Against a 190 AC target, I need to increase my to-hit by 20 POINTS to reduce my grazing hit roll by 1.

Or, I can reduce my targets AC by 20 points to achieve the same effect. This is already after using -15 for improved destruction.

How to close this gap?

1. Acquire Precision: This automatically gives 1 extra die roll since it provides 5% to-hit. Should offensive melees do more damage by foregoing power attack and associated enhancements like momentum swing and lay waste? This is a very counterproductive way to encourage more damage

2. Be an elf and acquire the 4-6% available in their cores. Why are such bonuses not present in the kensei tree, which is arguably the most to-hit centered tree in the game?

In summary, salient points:

1. Maxed out to-hit melees are missing 1/4 of their attacks on end-game bosses.
2. The amount of to-hit necessary to graze on 1 die lower is nearly unachievable when sitting at an almost 180 to-hit.
3. The only way to close this gap is to forego crucial melee attributes like power attack and LD abilities to use precision or to pidgeon hole into being an elven race.
4. Why does the Kensei tree not offer direct improvements to % in landing attacks, as they are the most to-hit oriented class in the game.

[Qhualor]

I wonder what the difference is between elite and R10? the reason I ask is because if its essentially just as bad, than I agree its a problem for melees and something needs to be done. if it's not, than I would see it as a Reaper problem and anything that should be done should be done Reaper related like the trees or Reaper mobs or something that wont power creep melees elite and below. I cant really agree or disagree as a general melee problem either way until there is an elite comparison.

[blerkington]

Yep, I don't think Turbine/SSG did us any favours when they started using a formula that makes it easier for poorly built and geared characters to hit and harder for people who really invested in this. But as I remember it that was actually mentioned as a plus when the new system was introduced.

Perhaps the to hit formula or how monster ACs are determined could be adjusted. It does seem like we are losing an unreasonably large amount of damage because of the way these two things are interacting, particularly on characters that are highly invested in accuracy.

Thanks.

[Renvar]

Yep, I don't think Turbine/SSG did us any favours when they started using a formula that makes it easier for poorly built and geared characters to hit and harder for people who really invested in this. But as I remember it that was actually mentioned as a plus when the new system was introduced.

Perhaps the to hit formula or how monster ACs are determined could be adjusted. It does seem like we are losing an unreasonably large amount of damage because of the way these two things are interacting, particularly on characters that are highly invested in accuracy.

Thanks.

Another way to look at this is that it front loads the benefits which limits the impact of PL's and such, making these systems more optional than they would be if it were linear. Also, it allows for more multi-classing options and build diversity to remain effective.

If you have to stay linear, then you are limited in range and it becomes nearly impossible to challenge the players playing 60 hours a week for years along with the newer ones or the more casual ones. A choice had to be made if you wanted to serve the meta-gamer/power gamer/grinder crowd and the rest of the customer base. Honestly, you could argue they should have let the metagamers either just roll alts or get bored and leave and keep the focus on the D&D PnP crowd. But that ship has sailed. And this is where we are.

EDIT: I don't think melee should be no-fail on To-Hit in R10. R10 is supposed to be a challenge, not an easy button.

[slarden]

I partly agree with the issue, but partly think recent demands for more power to complete R10s is a disturbing trend we've seen before during the process of watering down elite.

I think the main fixes needed for melee belong in the dread adversary tree - namely a cc clicky that works on reapers, a defense boost clicky at the top tier so it's something unique for melees dealing with pockets of reapers and champs. Also the ability to spend reaper points on % to hit chance would be a nice boost.

When the mechanic was new and before stealth changes I did exactly what you suggest, made a sun elf int assassin with precision + elven precision.

Couldn't you just turn on precision for that fight?

[redoubt]

Question: Do you think a nearly maxed to-hit fighter should have a 75% success rate in landing their melee attacks on end-game bosses? Here's the way things are currently working for me:

GRAZING HITS

Grazing hit formula as it stands:
(To-hit bonus + 10.5) / (2 * Target AC) + 25% for proficiency = % chance to hit rounded to the nearest 5% (this is determined by the die roll you see in your combat log).

R10 Black and Blue end dragon boss with a full improved destruction (-15 to AC):

Rolled a 5 with a 176 to-hit: Graze

How to close this gap?

1. Acquire Precision: This automatically gives 1 extra die roll since it provides 5% to-hit. Should offensive melees do more damage by foregoing power attack and associated enhancements like momentum swing and lay waste? This is a very counterproductive way to encourage more damage

2. Be an elf and acquire the 4-6% available in their cores. Why are such bonuses not present in the kensei tree, which is arguably the most to-hit centered tree in the game?

In summary, salient points:

1. Maxed out to-hit melees are missing 1/4 of their attacks on end-game bosses.
2. The amount of to-hit necessary to graze on 1 die lower is nearly unachievable when sitting at an almost 180 to-hit.
3. The only way to close this gap is to forego crucial melee attributes like power attack and LD abilities to use precision or to pidgeon hole into being an elven race.
4. Why does the Kensei tree not offer direct improvements to % in landing attacks, as they are the most to-hit oriented class in the game.

So...

1. I hate the departure from the D20. I hate the new formula. That said, it was not going to work on a d20 because the spread of "to-hit" between characters simply became too wide. (But still hate it.)

2. R10 boss dragon. You miss only 25% of the time. That sounds pretty awesome! In fact, you might need to be nerfed. (On elite I would expect a character like yours to hit 95% of the time.)

3. I like several of your ideas to make melee improvements. I think we simply disagree on the end point of those. For example, what if we targeted a maxed out fighter at 100% success on elite, then targeted mob AC to drop the success chance by 5% per skull? (For comparison, the current data on the wiki shows mob saves are +1.5 per skull, which on a d20 is 7.5% per skull.)

[Kaboom2112]

The ACs are too damned high!

Cetus is right, it's just dumb at this point. Lower them.

[Niminae]

I wonder what the difference is between elite and R10? the reason I ask is because if its essentially just as bad, than I agree its a problem for melees and something needs to be done. if it's not, than I would see it as a Reaper problem and anything that should be done should be done Reaper related like the trees or Reaper mobs or something that wont power creep melees elite and below. I cant really agree or disagree as a general melee problem either way until there is an elite comparison.

This would be great, until you remember that the devs have already said that whatever they decide to do for melees, it will not include any changes to the reaper trees. This despite the fact that the melee/physical DPs reaper trees are sadly lacking in comparison to the caster reaper tree.

Whatever we end up doing probably would not be linked to investment in Reaper trees, for the reason that we would want it to be a more universal improvement for melee in Reaper difficulty.

Does anyone think that melee need any help at all in Elite (or Hard, Normal, Casual) difficulty content?

[blerkington]

Another way to look at this is that it front loads the benefits which limits the impact of PL's and such, making these systems more optional than they would be if it were linear. Also, it allows for more multi-classing options and build diversity to remain effective.

If you have to stay linear, then you are limited in range and it becomes nearly impossible to challenge the players playing 60 hours a week for years along with the newer ones or the more casual ones. A choice had to be made if you wanted to serve the meta-gamer/power gamer/grinder crowd and the rest of the customer base. Honestly, you could argue they should have let the metagamers either just roll alts or get bored and leave and keep the focus on the D&D PnP crowd. But that ship has sailed. And this is where we are.

EDIT: I don't think melee should be no-fail on To-Hit in R10. R10 is supposed to be a challenge, not an easy button.

I think we could improve what we have without returning to a d20 system. As I said, they could look at tweaking the current formula or how monster ACs are assigned.

Our to-hit score is one area where there's not a large difference between characters with few past lives and those with many. There's really just fighter past lives and whatever you can get from stat increases, which makes little difference.

DPS loss through a significant amount of grazing hits occurs under r10 difficulty too. This problem is not confined just to the very hardest difficulty.

Thanks.

[Cetus]

Just to add: this isn't an R10 issue, I just happened to have shown data from an R10 boss.

In fact, R1 harry has a higher AC than that R10 boss dragon.

[zehnvhex]

My immediate knee-jerk reaction is sort of a "It's not like melee aren't king of the hill right now in DPS town, why exactly do you want to make it even better?" That line of reasoning though is single focused and leads to shenanigans like not boosting ranged damage because otherwise what's the point of melee, etc...

If you look at DDO as a whole then you'd could probably make a case for something along the lines of:

- Auto-attack damage is fine to have such high graze/miss rates
- Special attacks however should not be subject to the same miss rates

This would allow you to focus on a more ability heavy rotation which is better for bridging the gap between veteran and new players at the same time. With a more consistent/reliable source of damage, they could then nerf melee damage at cap to be more in line with what pure damage ranged/casters can do (or boost ranged/caster DPS to compensate).

You then give melee more survivability tools when they aren't tanking and not doing stupid things (IE: DPSing from the front) to compensate for the otherwise increased risk of being in melee range.

[Sam-u-r-eye]

This has been around since long before reaper. The to-hit values in legendary were very high.
That said...

...melee are king of dps atm. I think two-handers could use a little more love than Chism but your DPS is fine.

The problem of to-hit is worse on ranged than melee so any buff to melee to hit would have to be local. THAT said, a lot of ranged dps styles have fallen way behind (e.x., bows and repeaters) and the new raid shuriken is amazingly broken with how the bleed dot interacts like the old LE shroud boom sticks.

TL;DR

Why do you need more dps?

[Blastyswa]

This would be great, until you remember that the devs have already said that whatever they decide to do for melees, it will not include any changes to the reaper trees. This despite the fact that the melee/physical DPs reaper trees are sadly lacking in comparison to the caster reaper tree.



Does anyone think that melee need any help at all in Elite (or Hard, Normal, Casual) difficulty content?

I interpreted that second quote that's from Cordovan differently- I don't think it's a factor of "Any changes to to-hits for melee will be done in a way that affects all difficulties" and more of a "Any changes to to-hits for melee will be done in a way that doesn't make it only possible to be effective with Reaper xp". For example, they could add +5 to hit in each Adversary core and +1/2/3 to each of the MP and RP bonuses and add a passive +60 to-hit in reaper, which based on the original numbers would make a very significant difference for people with 40+ reaper points in adversary. Which then means you need 40+ reaper points in adversary to avoid the problem of a 75% hit rate. He could have meant something else, but my take on it was that any change to to-hits will involve either a change in the formula used, or a change in enemy AC.

[Cetus]

Actually, I don't mind having a project like this to work on.

It's going to be interesting for me and those in my grouping circle to put our heads together and see how much we can debuff a bosses AC or increase the to-hit of the party so as to graze as infrequently as possible.

I'm just pointing out the nature of the current functionality - you need astronomically high amounts of to-hit bonus to cross the threshold into 1 less graze in the d20.

Incremental bonuses no longer work, which sucks. Same goes for high reaper, incremental bonuses to damage like "+1 damage" may mean nothing if it's discounted by 95+% by stepping into R10.

I like the challenge, I just don't like the fact that gaining a, say, +10 to-hit can effectively mean nothing if I end up getting rounded down. And +10 is no joke of a difference at that level in terms of investment, let alone +20...

My essential point here is:

Yes to challenge.

No to large investments that mean nothing.

[Vish]

Bud, you're bouncing off the ceiling

I would say it sounds about right, r10 boss, 176 to hit it 75%
Yep, about as maxxed as it gets and youve reached the cap
So diminishing returns says at some point zero
You've got some to go but can't get there with whats designed
So, good design?
Like balanced, or that you're not at 95%
Because where would be the challenge then?
And pa and precision. There's your build choices
What you want it all pigeon holed?
I would say its all WAI
Hope you are getting your entertainment value

[blerkington]

This has been around since long before reaper. The to-hit values in legendary were very high.
That said...

...melee are king of dps atm. I think two-handers could use a little more love than Chism but your DPS is fine.

The problem of to-hit is worse on ranged than melee so any buff to melee to hit would have to be local. THAT said, a lot of ranged dps styles have fallen way behind (e.x., bows and repeaters) and the new raid shuriken is amazingly broken with how the bleed dot interacts like the old LE shroud boom sticks.

TL;DR

Why do you need more dps?

I'd actually prefer bosses to have more HP and for us to hit them more reliably.

I don't like grazing this much at cap with all the good gear and a decent build; it makes my character feel incompetent when he should be at his strongest. It also causes problems with how often special attacks hit like you said.

But if they made an adjustment, which would also help ranged players, they'd have to look at what increasing mob HP meant for nuking builds. I wouldn't want one (or two) playstyle(s) to gain power at the expense of others, nor do I want content to become easier.

Thanks.

[btolson]

Incremental bonuses no longer work, which sucks. Same goes for high reaper, incremental bonuses to damage like "+1 damage" may mean nothing if it's discounted by 95+% by stepping into R10.

...

No to large investments that mean nothing.

Balance aside, this is the crux of the problem - to-hit bonuses have not been dealt with well since the overhaul of the system so long ago. Look at Kensei tree - those and similar enhancements give out +1/+1 to hit/dam, +1/+2, as high as +1/+4, as if 1 point of to-hit was worth substantially more than even +4 dam. Before the overhaul this was may have been reasonable, since that meant +5%. But after? Those enhancements should all have given +1% to hit. The Bless spell should've given at least +1% to-hit, or else +1 to-hit per 2 caster levels. Divine Favor should've given +1% per 3 levels, or had its level cap removed, or a combination of the two. Etc, etc. None of these things were corrected when +1 to-hit got nerfed way hard in terms of its relative value.

That said, that's way more work than the devs are ever going to do. Better idea is to overhaul the system again, but this time use a formula where +1 to-hit always means something. +to-hit from items sorely needs to be brought in line with +to-hit from literally all other sources though (i.e., massive nerf), because those values are just ridiculously out of whack.

[Rys]

This problem is not limited to melee. It also influences ranged DPS. Especially if relying on FotW, the DPS loss is huge.

[AbyssalMage]

As far as I know, your calculations show that the To-Hit formula is WAI. It's been this way for quite some time (i.e since the implementation). I'm sure someone can dig up the 20+ page threads with your exact same experience.

[Cetus]

As far as I know, your calculations show that the To-Hit formula is WAI. It's been this way for quite some time (i.e since the implementation). I'm sure someone can dig up the 20+ page threads with your exact same experience.

Yes, the formula is WAI.

The intention is what I'm trying to get re-evaluated here.

Just like melee/spell/ranged power was re-evaluated, just as melee alacrity was re-evaluated with the introduction of doublestrike, just like PRR was re-evaluated, just like DR needs re-evaluation, to-hit needs evaluation as well.

Maybe we need something like "attack power" now? Let the D20 role be confined to dictating crits on weapons.

I don't have a good suggestion yet, I just want the devs to recognize that this is actually a problem right now. This rounding functionality is taking a huge dump on moderate investments to our attack.

Melee need an overhaul, this is an integral part of it.