The big melee DPS issue: Grazing Hits

[Fivetigers33]

Old System: Most of the time buffs did absolutely nothing, but they were game breaking a few combats.

New System: Buffs don't do as much, but they matter at least a little in most combats outside of the most extreme examples.

Attack bonus buffs and AC debuffs are currently virtually useless, except in those rare instances where adding a few to your attack bonus will push you to the next threshold. I did some Legendary Normal testing a while back and saw similar results to what Cetus is experiencing on R10 where squeezing out an extra +10 to attack or using improved destruction to lower enemy AC by 15 had no effect on whether a hit was grazing or not.

[Gremmlynn]

pft. sounds like BS.

A stars aligned melee build is only hitting 75% ..
Where's the investment payoff where building to become stars aligned actually means something.


DC casters have sweet spot thresholds to build towards.. why screw the melees.

IMO, DC casters are somewhat broken due to this (though, if changed to something closer to the weapon formula, some sort of grazing hit or SP discount when a save is made probably be be implemented).

What needs to be done is to fix the rounding to allow some sort of improvement to those who so choose to invest that way or who achieve the chaser items/character improvements that put them at those values.

[Gremmlynn]

Attack bonus buffs and AC debuffs are currently virtually useless, except in those rare instances where adding a few to your attack bonus will push you to the next threshold. I did some Legendary Normal testing a while back and saw similar results to what Cetus is experiencing on R10 where squeezing out an extra +10 to attack or using improved destruction to lower enemy AC by 15 had no effect on whether a hit was grazing or not.

Try it on a first life poor build in heroic content and you will see the opposite. Which is the point of the mechanic, it makes it easy to be viable, fairly smooth to become good, hard to be great and nearly impossible to become perfect (or more than perfect) within the mechanics.

[redoubt]

This current implementation is nowhere near "objectively better" than what it used to be. I disagree that because melee were hitting less that means the issue is worse. Heres why:

Old way - If my character was hitting far less, every single +1 I can muster means I hit 5% more often. If I trade off a few AC points, a point or two of saves, and a few HP to get +6 more to-hit that means I hit 30% more often.

New Way - Every single +1 does not alter the result - I can trade off a few AC points, a point or two of saves, and a few HP to get ABSOLUTELY NO DIFFERENCE in to-hit results.

If the logic is that hitting less often means the problem is worse, then all they needed to do to "fix" the issue in your example is lower Malicia's AC somewhat, if scoring hits on anything other than a 20 was impossible, or if it was possible (which is what I remember) it was on the player to do things to get better to-hit.

**Also** this Malicia example was in an era when buffs mattered and buff stacking was used alot more often to attain to-hit scores that worked. We stacked bard songs on top of multiple buffs and turned power attack off when needed in order to hit mobs more often. What are those buffs doing for you nowdays? Nothing - as the end result of the formula gets rounded down and you still hit the same number of times you did without them.

Old system: When buffs mattered

New system: Buffs dont matter

Still believe things are alot better now than back then?

While I'm okay with missing 25% of the time at end game R10, I'm with you and Cetus that investments in skills need to matter. When your +1s get lost in the formulas (or +10 gets lost in a formula) something is wrong. Both to hit and to damage need to be looked at in this regard.

As far as making it hard to hit and making buffs matter... Tempest spine on normal back when it was end game. Fighters would make a +5 sword. That's it. It was +5 of nothing and it was a quest to make it. The party would put every buff we could on the fighters and they were the only ones hitting the blackguards...

[Chai]

I do.

Personally, I see the biggest issue with rounding to the nearest 5%, for players, to pretend we are still using a d20 system (they don't do that with mobs). If, instead of rounding to 5%, they rounded to 5 decimal places of a percent it would cause every +1 to add something to a large degree.

With no rounding this would be correct. With rounding the same issue occurs.

In that case using the same formula (additive) with more scaling gradients is a far "objectively better" solution than changing the entire formula from an additive to parabolic.

It would just add relatively more to those who start with less. Which to my mind makes sense as the better one is at something, the harder it is to improve.

Then striving for mediocrity is the objective of the melee game. Once you hit that wall where the diminished returns are relatively negligible it becomes better to invest in something else than it is to continue investing in the same thing.

The main issue with this, is this game is then a model of inconsistency. To-hit and AC are now things you strive for mediocrity on, while saves and DC are still additive in nature. This is in part where power creep comes from in game. If I can trade away some to-hit in order to gain some saves, I just traded away no loss in end result on one end, to get an absolute gain on the other end. Those who claim to be allergic to power creep missed the boat on this one.

[Fivetigers33]

Try it on a first life poor build in heroic content and you will see the opposite. Which is the point of the mechanic, it makes it easy to be viable, fairly smooth to become good, hard to be great and nearly impossible to become perfect (or more than perfect) within the mechanics.

I did my tests on a level 26 caster oriented Bard, not a maxed out melee build. AB was only in the 50s. Buffs/debuffs didn't help.

[redoubt]

I agree. Also not stated is all of melees special attacks and cc need to hit to work. A comparison so our caster friends could understand is its like every mob for a melee has sr and you cannot get no fail sr checks. I know for a fact if that happened to casters they would whine to no end. Heck the whine when they couldnt get no fail dcs was epic. To hit fixes are needed or special melee attacks need to hit boosts.

Cast time
Concentration checks or quicken spell
SR checks (harder to beef up SP than to hit)
Bug where your feet are out of range
Bug where mob is in front of you, but it doesn't cast because game thinks mob is behind you (or was at some point)
Deathward
Freedom of Movement
Energy resists
Reflex saves
Evasion

And now back to discussing grazing hits...

[Chai]

While I'm okay with missing 25% of the time at end game R10, I'm with you and Cetus that investments in skills need to matter. When your +1s get lost in the formulas (or +10 gets lost in a formula) something is wrong. Both to hit and to damage need to be looked at in this regard.

Yeah I agree. The trade off for the character that does hit on a 2 or better on the hardest to-hit mobs is they had to sacrifice survival stats to get there.

As far as making it hard to hit and making buffs matter... Tempest spine on normal back when it was end game. Fighters would make a +5 sword. That's it. It was +5 of nothing and it was a quest to make it. The party would put every buff we could on the fighters and they were the only ones hitting the blackguards...

Either this, or we all equipped shields and shield wall+fire walled the blackguards down. Then it was the AC buffs that mattered.

[Gremmlynn]

I did my tests on a level 26 caster oriented Bard, not a maxed out melee build. AB was only in the 50s. Buffs/debuffs didn't help.

That is due to the large AC value in the denominator and the rounding.

[redoubt]

By the way, even when you're "on the die" and the buffs actually mattered (which was rare), your numbers are still wrong. If you improve your To Hit so that you have an additional +6 of relevant attack value under the old system, you aren't hitting 30% more often. You're hitting for anywhere from 46% to 600% more often. Your math is terrible.

Okay, trying to work numbers with you.

50 needed to hit.

+39 to hit score. Need to roll 11 to hit. 50% hit rate.

+4 to hit from GH. Need to roll a 7. 70% hit rate.

New one, still 50 needed.

+29 to hit. Rolling a 20 is not enough. But you hit on a 20 anyway.

+4 from GH. Need to roll 17. 20% hit rate. In this case, because you started "off the die" and 20s were auto hits you only gained 15% chance to hit.

New one, 50 needed.

+47 to hit. Need to roll a 3. 85% hit rate.

+4 from GH. Don't need to roll to hit. Only gained 15% because you moved "off the die" top of scale.

Note: ignored auto miss in math.

As long as you were on the die, you gained and lost 5% per point of to hit.

Lots of stacking buffs / debuffs had the ability to put you "on the die".

[redoubt]

It seems like much of the issue comes in once we start rounding numbers. One fix suggested for it is to not round any more. This is plausible, but the small trade off is that we would end up scrapping seeing a die roll.

Another option that has been discussed in the past is a larger die. Is it time to discuss this again?

[HungarianRhapsody]

Okay, trying to work numbers with you.

50 needed to hit.

+39 to hit score. Need to roll 11 to hit. 50% hit rate.

+4 to hit from GH. Need to roll a 7. 70% hit rate.

New one, still 50 needed.

+29 to hit. Rolling a 20 is not enough. But you hit on a 20 anyway.

+4 from GH. Need to roll 17. 20% hit rate. In this case, because you started "off the die" and 20s were auto hits you only gained 15% chance to hit.

New one, 50 needed.

+47 to hit. Need to roll a 3. 85% hit rate.

+4 from GH. Don't need to roll to hit. Only gained 15% because you moved "off the die" top of scale.

Note: ignored auto miss in math.

As long as you were on the die, you gained and lost 5% per point of to hit.

Lots of stacking buffs / debuffs had the ability to put you "on the die".

Don't just look at the percentage scores - look at the number of hits. Each "5% increase" doesn't mean 5% more hits. It means a lot more than that, depending on where you are on the die. Or it means no change at all if you're off the die.

If you are attacking 20 times or 400 times or however many times, that percent is the percent of those attacks that will land. If you're going from "hitting on a 20 only" to "hitting on a 17-20" as an example, you are now hitting four times as often. In 20 attacks, you were hitting an average of once. Now you are hitting an average of four times. In 400 attacks, you were hitting an average of 20 times and now you're landing an average of 80 hits. That "+15%" ended up quadrupling your hits because you previously were only hitting 5% of the time to begin with.

[Gremmlynn]

With no rounding this would be correct. With rounding the same issue occurs.

In that case using the same formula (additive) with more scaling gradients is a far "objectively better" solution than changing the entire formula from an additive to parabolic.

Except it's not parabolic. at least the formula in the OP isn't (nor is it much more than 4th grade math, 8th grade pre-algibra is closer).

Then striving for mediocrity is the objective of the melee game. Once you hit that wall where the diminished returns are relatively negligible it becomes better to invest in something else than it is to continue investing in the same thing.

Well the flat +.2 for proficiency keeps it ahead of mediocrity to start (not being proficient at something, pretty much means mediocre at best). The thing it does is allow characters in various levels of progression to play the same game.

The main issue with this, is this game is then a model of inconsistency. To-hit and AC are now things you strive for mediocrity on, while saves and DC are still additive in nature. This is in part where power creep comes from in game. If I can trade away some to-hit in order to gain some saves, I just traded away no loss in end result on one end, to get an absolute gain on the other end. Those who claim to be allergic to power creep missed the boat on this one.

Agree there, the DC formula should be something like the "To Hit" formula.

Also, these sorts of formulas actually help curb the effects of power creep, though they could have been done better.

[Qhualor]

Cast time
Concentration checks or quicken spell
SR checks (harder to beef up SP than to hit)
Bug where your feet are out of range
Bug where mob is in front of you, but it doesn't cast because game thinks mob is behind you (or was at some point)
Deathward
Freedom of Movement
Energy resists
Reflex saves
Evasion

And now back to discussing grazing hits...

Not to be argumentative, but it's not much different from what a melee has to go through.

[Gremmlynn]

It seems like much of the issue comes in once we start rounding numbers. One fix suggested for it is to not round any more. This is plausible, but the small trade off is that we would end up scrapping seeing a die roll.

Another option that has been discussed in the past is a larger die. Is it time to discuss this again?

d1,000,000?

[HungarianRhapsody]

With no rounding this would be correct. With rounding the same issue occurs.

Just changing the rounding to "to the nearest percentage point" instead of "to the nearest 5%" would fix most of this issue.

Edit: and that's assuming this is a problem that needs a fix.

[Chai]

Yes. I still believe that things are a lot better now. Mostly because you're doing the math wrong by misunderstanding percentages.

This belief system doesnt trump logic.

Back then most To Hit buffs didn't matter either.

Simply not true. Back then they mattered and were used when necessary. Angog the champion, blackguards in TS, and Malicia are but a few examples. Posted builds often had to-hit and +damage breakdowns. There were several threads discussing if turning power attack off meant a DPS increase, and how far off the "always hit" mark you needed to be to do it or not. Grouping with bards, cleric recitation, greater heroism, were often sought after - and not just in a few specific examples - but rather in all of the game.

If you need a 70 To Hit and your To Hit was 40, that +1 or even +7 did absolutely nothing for you.

No one built melee that poorly where they were that far off the die without being able to receive easy-to-implement advice that would fix this. If they were that far off the die on the "needs improvement" side of things there were in game mechanics that were not being used yet by that player.

If you need a +50 To Hit and you have a 49 (or more) To Hit, then no buff of any kind had any meaning AT ALL for you. If you weren't "on the die", then the buffs didn't matter. And it was incredibly rare for someone to be "on the die".

This player in the posted unrealistic example didnt challenge themselves enough if they were in this situation. DDO provided enough situations where maxed out AB was still on the dice and within 20 increments.

Old System: Most of the time buffs did absolutely nothing, but they were game breaking a few combats.

It is sounding more and more like you did not play DDO from 2006 to 2011 or so. Buffs were highly sought after in all aspects of the game. Where do you think the term "axer package" came from. It was common DDO forum slang for a specific set of optimized buffs a specific poster always wanted.

New System: Buffs don't do as much, but they matter at least a little in most combats outside of the most extreme examples.

Also objectively wrong (incomplete at best). It is highly likely after reaching specific thresholds that reasonable increases which meant a sacrifice somewhere else will result in no gain in the end result. AKA you traded something for nothing.

[Chai]

By the way, even when you're "on the die" and the buffs actually mattered (which was rare), your numbers are still wrong.

Nope. My numbers are absolutely correct. Its blatantly obvious you started playing after most of the buffs were eroded out of the game (through your continued claims of how they rarely mattered). No one who played from 2006-2011 would make an objective claim that they rarely mattered. GO to archive.com and look at melee build threads in 2010 or so and see all the to-hit breakdowns included in them.

If you improve your To Hit so that you have an additional +6 of relevant attack value under the old system, you aren't hitting 30% more often. You're hitting for anywhere from 46% to 600% more often. Your math is terrible.

You are speaking in terms of increase in effectiveness comparing the before to the after only, with zero consideration for the entire field of possibilities.

I am speaking in terms of total effectiveness. On a 20 sided dice each increment represents 5% probability.

[Chai]

Just changing the rounding to "to the nearest percentage point" instead of "to the nearest 5%" would fix most of this issue.

Edit: and that's assuming this is a problem that needs a fix.

Or we could talk about how much content we could have received in lieu of all the effort that went into revamping a system to push the game away from its namesake.

Any claims that this system we use currently is due to popularity reasons will be refuted by a challenge to create a thread suggesting the same be done with saves and DC. Let me know when you are going to get that conversation going . I want to buy stock in torches and pitchforks the week before that. Redenbacher will split twice in the same week.

[Renvar]

New Way - Every single +1 does not alter the result - I can trade off a few AC points, a point or two of saves, and a few HP to get ABSOLUTELY NO DIFFERENCE in to-hit results.

Honestly, what you describe sounds like a poor builder making a bad build decision.

This has always been possible from the beginning of the game (and most any RPG I can think of).

Odd stats is the most common example. Which is better? 16 str, 12 dex, 14 con, 12 wis, 8 int, and 8 cha or 17 str, 11 dex, 14 con, 10 wis, 8 int, and 8 cha? I'm sacrificing dex and wis for strength, but the net result is... ABSOLUTELY NO DIFFERENCE in to-hit results.

You never want to reduce secondary stats/abilites just to get a primary stats or abilities increase that don't result in a numerical advantage. You ALWAYS want to manage your build trade offs to find these sweet spots. Whether the gap is 2 points or 20 points, making an investment that just moves you from the bottom to the middle of the same range is a poor build decision. Not a systemic flaw.

The system can absolutely be designed such that the incremental costs increase as you go up the power range, especially if you want a game that is trying to maintain a single character perpetual play environment and serve both power gamer and casual player base simultaneously. (Which also mirrors real life. As you move up the skill tree, it takes progressively more investment to improve incrementally) Also, maybe there are peaks that your players can't currently climb today, but will be able to in the future. Remember, the game is an ongoing story that has not been fully told. A journey not yet completed. They can only get so far with what has been released thus far. And the current max leaves them in the middle of a range. Smart players will realize this and take advantage of that to increase secondary stats/abilities with no decrease in their primary "max" stat/ability. When new gear/systems are released, you can ascertain if it will get you to the next statistically relevant point or not. And build accordingly.