[3.5] Diminishing Return Fumbles -- Common?

[Maginomicon]

A lot of you have problems with the notion of fumbles, for various reasons, and that's fine.

How about this though?


Every time you would fumble an attack roll, you must also roll to confirm that you missed. However, these confirmation rolls are against the creature's touch AC no matter what the original attack was against. Further, you must roll your confirmation roll a number of times equal to the number of iterative attacks you would normally gain from your Base Attack Bonus (so once with BAB +0, twice at BAB +6, three times at BAB +11, etc.). If you "hit" on any of those rolls, it resolves as a miss but not a fumble.

This solves two problems:

If your weapon at least contacted with the target, the attack can't fumble, period (although this is meaningless for attacks that were touch attacks in the first place).
Your chance of a fumble does diminish with experience (essentially into oblivion), and it does so for everyone (although more martial-oriented characters have it diminish faster).


Does this hold up to the 100-man straw target test?

Regardless, discuss. :smallwink:

NOTE: This is not a thread for whining about the notion of people using fumbles in their game. I also recognize that caster-types don't have something similar described here, but I don't care, as I've already implemented a "miss chance"-like effect to spellcasting (http://www.giantitp.com/forums/showthread.php?t=306961). Talking about those sorts of things is thus blatantly off-topic (and might even be thread-crapping).

[Cog]

Rolling an attack half a dozen times just to miss at higher levels sounds like a heck of a lot of wasted time to me.

[Baroknik]

I can't say whether or not it fully works mathematically at the moment (though later tonight I may have the time/capacity to do a rigorous modeling of the system). However, I will note that an interesting mechanic of this system is that targets that rely on "elusiveness" will force fumbles more often, which is interesting to me.

For example, picture a target with 100 dex (+45 modifier) and no armor. If you have a fumble chance against this person, then you have a very high chance of fumbing (do 20's on any of the iteratives auto-cancel out the fumble chance, or does the auto-hit 20 not apply on the fumble chance?). The opposite however, means that the guy wearing really thick slabs of metal (+45 armor for total 55 AC) is going to almost never be fumbled against (DC 10 to resist the fumble). Thematically, it fits pretty well that that would be the case (picture a monk or a rogue forcing a misstep vs a panzer tank trying to make you overreach), so I do like it as far as verisimilitude goes in that regard.

Also, what about miss chance from concealment? If the target has concealment (20% miss chance), do each one of the fumble recovery attempts apply to this as well? That is, if they have a touch AC of 15 and a 20% miss chance and on my first recovery attempt I roll a 16, but then roll a 05, do I recover or still miss?

[Dusk Eclipse]

This seems needlessly complicated to me, if you want to use fumbles at your table feel free to do so, but honestly this is just a time waster. D&D combat is already long enough without having to introduce unnecessary rolling.

[Xervous]

Viewing the average monster stats (http://www.giantitp.com/forums/showthread.php?t=172050), specifically their touch ACs... I took the average and max values and developed a linear relation with respect to level.

let x = level

Average touch trendline = -0.1425x + 11.579
Max touch trendline = -0.2865 + 19.858

*Note* there are some outlier values for max touch AC in excess of 20. Of course, expect the practicality of this rule to vary based on the nature of the enemies the players are fighting.

Seeing this, we may conclude that touch ACs are generally lower at higher levels, making it harder to confirm your critical fumbles.

Assuming an 18 strength fighter who obtains a minimum cumulative bonus of +1 to hit every two levels (an underestimation for our case). Thus, his standard attack bonus is 1.5(Level) + 4.

From our trendline, assume an average of around 12 for touch AC at level 1. The fighter has +5 to hit, so he has only a 35% chance of confirming his fumble, putting the overall chance of fumble per attack at 1.75%. A wizard of this level making a melee attack with a +0 bonus (generous) has a 60% chance of confirming his fumble, putting it at 3%.

Jump ahead to level 6, average touch is mostly 12 (though more like 11, used 12 for following calcs)). Fighter has +13 to hit, making his first attack a 1/8000 fumble 0.0125%... His second attack at +8 is a 1/500 fumble, 0.2%. Combined, this is a 0.212475% fumble chance. Wizard will have a cute +3 to hit, giving him a 2.25% fumble rate. Wow, now we see the wizard fumbling at least ten times for every one time the fighter fumbles, assuming they are both making the same number of full attacks.

Level 11, touch AC of 11, Fighter with +20/+15/+10, all attacks only confirm fumbles 1/160,000 times. 0.001875% chance of a single fumble per attack. Wizard has +5, giving him a 1.75% confirmed fumble rate.

Beyond this point, I don't feel it is necessary to flood the screen in zeroes to show that the fighter won't be confirming many fumbles.



Though, there is one area that draws my concern, that of two weapon fighting. Naturally, TWFers aren't going to be swinging around as high attack bonuses as THFs. Additionally, they are making more attacks and thus more likely to fumble.

Of course, I have said nothing about that practicality of implementing such a feature.

[Luciandevine]

This seems needlessly complicated to me, if you want to use fumbles at your table feel free to do so, but honestly this is just a time waster. D&D combat is already long enough without having to introduce unnecessary rolling.

I agree. If you want to use fumbles at your table, it's your choice as the dm to implement them. likewise, if you, as the dm don't like them, then simply don't use them.

I agree with Eclipse that the last thing that needs to happen with traditional combat is lengthening it with more dice-rolling whose only positive result is a normal miss instead of a fumble.

[Maginomicon]

I can't say whether or not it fully works mathematically at the moment (though later tonight I may have the time/capacity to do a rigorous modeling of the system). However, I will note that an interesting mechanic of this system is that targets that rely on "elusiveness" will force fumbles more often, which is interesting to me.

For example, picture a target with 100 dex (+45 modifier) and no armor. If you have a fumble chance against this person, then you have a very high chance of fumbing (do 20's on any of the iteratives auto-cancel out the fumble chance, or does the auto-hit 20 not apply on the fumble chance?). The opposite however, means that the guy wearing really thick slabs of metal (+45 armor for total 55 AC) is going to almost never be fumbled against (DC 10 to resist the fumble). Thematically, it fits pretty well that that would be the case (picture a monk or a rogue forcing a misstep vs a panzer tank trying to make you overreach), so I do like it as far as verisimilitude goes in that regard.

Also, what about miss chance from concealment? If the target has concealment (20% miss chance), do each one of the fumble recovery attempts apply to this as well? That is, if they have a touch AC of 15 and a 20% miss chance and on my first recovery attempt I roll a 16, but then roll a 05, do I recover or still miss?

I love the idea of high-DEX characters being able to subtly manipulate an attacker into having more fumbles. Awesome.


Viewing the average monster stats (http://www.giantitp.com/forums/showthread.php?t=172050), specifically their touch ACs... I took the average and max values and developed a linear relation with respect to level.

let x = level

Average touch trendline = -0.1425x + 11.579
Max touch trendline = -0.2865 + 19.858

*Note* there are some outlier values for max touch AC in excess of 20. Of course, expect the practicality of this rule to vary based on the nature of the enemies the players are fighting.

Seeing this, we may conclude that touch ACs are generally lower at higher levels, making it harder to confirm your critical fumbles.

Assuming an 18 strength fighter who obtains a minimum cumulative bonus of +1 to hit every two levels (an underestimation for our case). Thus, his standard attack bonus is 1.5(Level) + 4.

From our trendline, assume an average of around 12 for touch AC at level 1. The fighter has +5 to hit, so he has only a 35% chance of confirming his fumble, putting the overall chance of fumble per attack at 1.75%. A wizard of this level making a melee attack with a +0 bonus (generous) has a 60% chance of confirming his fumble, putting it at 3%.

Jump ahead to level 6, average touch is mostly 12 (though more like 11, used 12 for following calcs)). Fighter has +13 to hit, making his first attack a 1/8000 fumble 0.0125%... His second attack at +8 is a 1/500 fumble, 0.2%. Combined, this is a 0.212475% fumble chance. Wizard will have a cute +3 to hit, giving him a 2.25% fumble rate. Wow, now we see the wizard fumbling at least ten times for every one time the fighter fumbles, assuming they are both making the same number of full attacks.

Level 11, touch AC of 11, Fighter with +20/+15/+10, all attacks only confirm fumbles 1/160,000 times. 0.001875% chance of a single fumble per attack. Wizard has +5, giving him a 1.75% confirmed fumble rate.

Beyond this point, I don't feel it is necessary to flood the screen in zeroes to show that the fighter won't be confirming many fumbles.

Though, there is one area that draws my concern, that of two weapon fighting. Naturally, TWFers aren't going to be swinging around as high attack bonuses as THFs. Additionally, they are making more attacks and thus more likely to fumble.

Of course, I have said nothing about that practicality of implementing such a feature.
Thank you. That was much more helpful than the other responses so-far.


Rolling an attack half a dozen times just to miss at higher levels sounds like a heck of a lot of wasted time to me.

This seems needlessly complicated to me, if you want to use fumbles at your table feel free to do so, but honestly this is just a time waster. D&D combat is already long enough without having to introduce unnecessary rolling.

I agree. If you want to use fumbles at your table, it's your choice as the dm to implement them. likewise, if you, as the dm don't like them, then simply don't use them.

I agree with Eclipse that the last thing that needs to happen with traditional combat is lengthening it with more dice-rolling whose only positive result is a normal miss instead of a fumble.
Annnnd the thread-crapping (not to mention ignorance) begins. Some of you clearly didn't care to really read the OP.

Here's a breakdown for you since you apparently need it spelled out:

1 extra roll 5% of the time at 1st level.
1 or 2 extra rolls 5% of the time at 6th level (more likely 1 since you're more and more likely to have hit the Touch AC once with each attempt, and once you "hit" once there's no need to make another roll).
1 or 2 or 3 extra rolls 5% of the time at 11th level (ditto regarding obviating rolls).
1 or 2 or 3 or 4 extra rolls 5% of the time at 16th level (ditto regarding obviating rolls).



Half a dozen extra rolls? Don't make me laugh. If anything it's a third of a dozen extra rolls, tops. (and not even that since "hitting" on any confirmation roll obviates any further rolling).

And the 5% thing? That can be made moot really easily by using the Bell Curve Rolls variant (http://www.d20srd.org/srd/variant/adventuring/bellCurveRolls.htm) instead of the standard d20, as the chance of rolling a natural 3 is "1 in 216" or 0.46% instead of a d20's 5%.

Why don't you just say "This seems needlessly complicated to me, if you want to roll initiative at your table feel free to do so, but honestly this is just a time waster. D&D narrative is already long enough without having to introduce unnecessary combat."

[eggynack]

Why do you need to roll several times in the confirmation scenario at all? If one of the dice comes up as a hit, then it doesn't matter what the other rolls are, and order doesn't matter. Just have them all be rolled simultaneously. As for the initiative thing, people often seek out solutions for how roll and complexity inducing that system is, so your sarcasm isn't far from the truth. You're applying complexity of approximately that scale to something which is very much not necessary for the game, and it is generally a poor decision to spend a game's complexity and time in unnecessary ways. Adding complexity is a decision that should not be taken lightly.

[Maginomicon]

Regarding saying that the premise is a waste of time...
Also, comments that are pointless negation of the purpose of a thread ("threadcrapping") or comments predicting the thread will end badly ("doomsaying") are also likely to be trolling.Everyone, please, don't do something that would influence me to start reporting people.

Besides, the fact that the extra roll or two happens a nearly-(or increasingly)-negligible percentage of the time further makes it even MORE silly to complain about it.


Why do you need to roll several times in the confirmation scenario at all? If one of the dice comes up as a hit, then it doesn't matter what the other rolls are, and order doesn't matter. Just have them all be rolled simultaneously. As for the initiative thing, people often seek out solutions for how roll and complexity inducing that system is, so your sarcasm isn't far from the truth. You're applying complexity of approximately that scale to something was is very much not necessary for the game, and it is generally a poor decision to spend a game's complexity and time in unnecessary ways. Adding complexity is a decision that should not be taken lightly.
Rolling all of them at the same time is an excellent idea, but dunno how many d20s a player's going to have on hand. Still, it's already suggested somewhere that you roll to-hit and damage rolls at the same time, so simultaneous rolls aren't unheard-of. All that matters is the highest d20 roll of the bunch, so yeah.

[Carth]

The two most common complaints about fumbles are that they're not fun, and a waste of precious table time. I get that you don't want the hate train running over your idea, but what are you hoping to accomplish with it? If you want feedback on how to improve your idea, having a more precise goal would help, we need to know what you're trying to improve. You can report myself and others if you want, but I have a hard time believing that the moderators are going to be upset that people are trying to talk you out of what they honestly see as a bad idea.

[The Trickster]

I know someone posted this already, but are % miss chances calculated during the new reroll?

[Maginomicon]

I know someone posted this already, but are % miss chances calculated during the new reroll?
I can see why it might be important, but I would say no. You roll % miss chance before your attack roll anyway because if you miss with your % miss chance there's no need to make an attack roll that could fumble in the first place (that is, with a % miss chance you utterly whiff by swinging in the wrong place, and so there's no reason why you could fumble).

[The Trickster]

The two most common complaints about fumbles are that they're not fun, and a waste of precious table time. I get that you don't want the hate train running over your idea, but what are you hoping to accomplish with it? If you want feedback on how to improve your idea, having a more precise goal would help, we need to know what you're trying to improve. You can report myself and others if you want, but I have a hard time believing that the moderators are going to be upset that people are trying to talk you out of what they honestly see as a bad idea.

I think the point of this was to implement a critical miss chance system that (basically) eliminates the possibility of a high level fighter (or whatever martial class) from fumbling. Having a near epic level character randomly tripping over his own shoelaces was always seen as a major problem with many critical fail systems. This idea aims to eliminate this issue, an it seems to mathmatically solve that problem.

[eggynack]

I think the real math/mechanical problem with this system comes in at the 1-5 range. A fifth level fighter is presumed to be at approximately the maximum of human capability, and he's fumbling at least as much as he would in the systems which are commonly disparaged. A fumble is going to occur at least once in every 400 attacks, and that's rather high. At those levels, it seems to fall under the Philistine chair hitting region of fumbling potential. To quote him directly:

Critical Fumble Rule:
If at any time a DM shall propose using a "critical failure" or "fumble" table of any sort in a 3.X game, the players are to beat the DM with folding chairs until each of them has accidentally struck himself with his chair at least once, while keeping a count of the number of strikes made before this happens. Then, the average rate of such "fumbles" as generated by a table full of nerds swinging improvised weapons will establish the maximum probability of a "fumble" within the game mechanics for a level 1 Commoner (note that this already will probably require rolling multiple Natural 1's in succession to confirm a fumble), with the probability dropping by at least an order of magnitude per point of BAB of the attacking character. Thus a full-BAB character at level 20 might have to roll 20+ Natural 1's in a row to before you even bother glancing at the Fumble Table.
At level six, the percentage is going to drop off drastically, but the fifth level thing seems problematic to me.

Edit: Actually, forgot a thing. What happens on a critical fumble? I might have missed it, but it doesn't seem to be in the rule as presented.

[Maginomicon]

At level six, the percentage is going to drop off drastically, but the fifth level thing seems problematic to me.

Edit: Actually, forgot a thing. What happens on a critical fumble? I might have missed it, but it doesn't seem to be in the rule as presented.
Nothing. All that matters on the confirmation is whether it missed or hit, just like rolling a natural 20 on a critical hit confirmation roll does nothing special.

Alternatively, you could use that as reasoning for using Dragon Compendium's "Bad Misses" fumble table (be sure to read the errata though).

[eggynack]

Nothing. All that matters on the confirmation is whether it missed or hit, just like rolling a natural 20 on a critical hit confirmation roll does nothing special.

Wait, sorry, meant what happens on a fumble in general? Does the guy drop his weapon, or hit himself in the face, or do random minor magical and pointless magical effects occur, or what? I like the last one. No one expects the fighter's sword to blink morse code in octarine, before suddenly not doing that, causing nothing.

[Maginomicon]

Wait, sorry, meant what happens on a fumble in general? Does the guy drop his weapon, or hit himself in the face, or do random minor magical and pointless magical effects occur, or what? I like the last one. No one expects the fighter's sword to blink morse code in octarine, before suddenly not doing that, causing nothing.
I just figured you could use whatever fumble effect you like, although I personally would prefer the Dragon Compendium "Bad Misses" fumble table. It uses a d% to determine the end result, with the weaker effects being by-far more common by percentile.

[Slipperychicken]

Diepool systems like Shadowrun do a good job of making sure fumbles don't happen that often. So we could emulate that somewhat.

Here's one idea- whenever you miss on a natural 1, roll 1d6 for each point of BAB you have. If all the d6s come up as 1s (the minimum result), then you fumble. If you have zero BAB, then natural 1s are automatic fumbles. This makes sure that skilled warriors (i.e. with high BAB) fumble much less often, while unskilled characters fumble more.

You could apply this system to skills too, only with skill ranks determining the number of d6's instead of BAB. A supernaturally-skilled character will almost never fumble, while an untrained person will fumble more often.

[eggynack]

I just figured you could use whatever fumble effect you like, although I personally would prefer the Dragon Compendium "Bad Misses" fumble table. It uses a d% to determine the end result, with the weaker effects being by-far more common by percentile.
It's a pretty important factor, I think. If you don't know what happens, how do you know what to look for when you beat the DM with chairs? As for these rules in particular, they seem pretty highly detrimental. Assuming no shield, a bit less than 2/3's of the results have checks, which is a good thing. However, some of those save results are basically ultra-deadly, like the oddly doubly listed 1/5 chance of getting that 1d4 rounds of stunning result. That's just going to kill the character in question a good amount of the time. The check isn't particularly low either, with a character with a dex penalty just failing. About 1/4 of the effects are hitting yourself in the face, which fall under philistine's result. Offhand, I would call this fumbling result somewhat harsh, though I'd have to look deeper to make sure.

[Fax Celestis]

Rolling an attack half a dozen times just to miss at higher levels sounds like a heck of a lot of wasted time to me.

Agreed. While this system is functional, it is slow and clunky.

[KillianHawkeye]

Question: does using those extra rolls from BAB to avoid confirming the fumble use up your iterative attacks, or are they simply bonuses that represent advancing skill as a warrior?

[Maginomicon]

Question: does using those extra rolls from BAB to avoid confirming the fumble use up your iterative attacks, or are they simply bonuses that represent advancing skill as a warrior?
It doesn't consume your iterative attacks. It simply progresses at the same rate as your iteratives and has nothing to do with iteratives or BAB otherwise.

[TuggyNE]

Using the averages from Optimization by the Numbers (http://www.minmaxboards.com/index.php?topic=3472.msg44888#msg44888), here's a summary by level of fumbles, with some vaguely-suitable attack bonuses assumed*:

{table=head]Level|AB|AC|TAC|Chance to miss or fumble|Chance to fumble
1|+5|15|12|45%|1.50%
2|+7|16|12|45%|1.00%
3|+8|16|12|35%|0.75%
4|+10|16|11|25%|0.25%
5|+11|17|11|25%|0.25%
6|+13|19|11|25%|0.01%/0.02%
11|+20|24|11|15%|~0%
16|+28|32|10|15%|~0%[/table]

This seems workable, although high touch AC could make things very funky. Actually, I'd suggest just giving a single extra confirmation roll at BAB +6 and calling it good; the others don't really seem necessary unless you're using a fumble table that's too harsh at low levels.

*If anything, these are likely to be a bit on the low side.

[Spuddles]

Viewing the average monster stats (http://www.giantitp.com/forums/showthread.php?t=172050), specifically their touch ACs... I took the average and max values and developed a linear relation with respect to level.

let x = level

Average touch trendline = -0.1425x + 11.579
Max touch trendline = -0.2865 + 19.858

*Note* there are some outlier values for max touch AC in excess of 20. Of course, expect the practicality of this rule to vary based on the nature of the enemies the players are fighting.

Seeing this, we may conclude that touch ACs are generally lower at higher levels, making it harder to confirm your critical fumbles.

Assuming an 18 strength fighter who obtains a minimum cumulative bonus of +1 to hit every two levels (an underestimation for our case). Thus, his standard attack bonus is 1.5(Level) + 4.

From our trendline, assume an average of around 12 for touch AC at level 1. The fighter has +5 to hit, so he has only a 35% chance of confirming his fumble, putting the overall chance of fumble per attack at 1.75%. A wizard of this level making a melee attack with a +0 bonus (generous) has a 60% chance of confirming his fumble, putting it at 3%.

Jump ahead to level 6, average touch is mostly 12 (though more like 11, used 12 for following calcs)). Fighter has +13 to hit, making his first attack a 1/8000 fumble 0.0125%... His second attack at +8 is a 1/500 fumble, 0.2%. Combined, this is a 0.212475% fumble chance. Wizard will have a cute +3 to hit, giving him a 2.25% fumble rate. Wow, now we see the wizard fumbling at least ten times for every one time the fighter fumbles, assuming they are both making the same number of full attacks.

Level 11, touch AC of 11, Fighter with +20/+15/+10, all attacks only confirm fumbles 1/160,000 times. 0.001875% chance of a single fumble per attack. Wizard has +5, giving him a 1.75% confirmed fumble rate.

Beyond this point, I don't feel it is necessary to flood the screen in zeroes to show that the fighter won't be confirming many fumbles.



Though, there is one area that draws my concern, that of two weapon fighting. Naturally, TWFers aren't going to be swinging around as high attack bonuses as THFs. Additionally, they are making more attacks and thus more likely to fumble.

Of course, I have said nothing about that practicality of implementing such a feature.

People like you are why I love this community :smallbiggrin: