Is the Critical Miss rule good for anything?

[Larkas]

I'm not talking about any critical fumble rules here, I'm talking about the core rule of always missing an attack or failing a save on a natural roll of 1. I can see the symmetry here: 20 always hits, 1 always misses. But is that even a good argument for that rule? I mean, regardless of training, you can't shake off that 5% chance of missing the attack or failing that save. You can be the greatest swordsman or the wisest sage ever to live, you still have that ridiculous 1-in-20 chance of failing due to pure, dumb bad luck. A critical hit makes the game more cinematographic, a critical miss doesn't. Even if the miss was on the part of a villain. At least not at the rates established by the "1-in-20" thing. What's worse, that rule penalizes harder the characters that roll more dice, which means the characters that have higher BAB, generally. If the character is getting better at hitting something, why does he have a higher statistical chance of automatically missing than someone who's worse at hitting something?

I don't think this rule is good for anything at all. The game doesn't need the symmetry suggested by it. I can see two ways out:

1 - Do away with the critical miss rule entirely. Maybe introduce a flaw for games with those ("Unlucky: You miss attacks and fail saves on a natural roll of 1").

2 - Make it so critical misses must be confirmed, that is, if you roll a natural 1, you have to roll the d20 again, and only automatically misses on another natural 1. Otherwise, the first roll is treated as "just 1". This drives the automatic miss chance all the way down to 0,25%, much more reasonable and manageable.

Using the stock rules:

- a fighter with four attacks will automatically miss at least one of them (any of them, including the one at the highest BAB) 18.55% of the time;

- a two-weapon fighting ranger with 7 attacks will automatically miss at least one of them 30.17% of the time;

- a monk flurrying with two-weapon fighting and haste with 9 attacks will automatically miss at least one of them 36.97% of the time.

Using the "two consecutive 1s" rule:

- a fighter with four attacks will automatically miss at least one of them 1% of the time;

- a two-weapon fighting ranger with 7 attacks will automatically miss at least one of them 1.74% of the time;

- a monk flurrying with two-weapon fighting and haste with 9 attacks will automatically miss at least one of them 2.23% of the time.

So, what do you guys think? Is it really a poorly thought-out and harmful rule? Does it have any pros I'm not seeing?

[zephyrkinetic]

2 - Make it so critical misses must be confirmed, that is, if you roll a natural 1, you have to roll the d20 again, and only automatically misses on another natural 1. Otherwise, the first roll is treated as "just 1". This drives the automatic miss chance all the way down to 0,25%, much more reasonable and manageable.

My 2 cp:

In my games, a Nat 20 is a crit threat. Afer that, you must roll again. Hitting the target AC is a confirmed crit, not doing so is still a regular hit. Another 20 is a insta-kill threat, after which you roll a third time, and the same rules apply.

So, for a Nat 1, call it a fail threat. After that, roll again. Hitting the target AC, you're ok; call it a "near miss," and damage is halved. Miss the target AC on the second roll, (but don't roll a 1), and you've got a miss on your hands. Two 1s and another fail, you're into broken-weapon/stabbed your friend territory.

To illustrate: (with -> indicating a die roll)
-> Meet or exceed AC = Hit
-> Natural 20 -> Miss AC = Hit
-> Natural 20 -> Meet or Exceed AC = Critical Hit
-> Natural 20 -> Natural 20 -> Miss AC = Critical Hit
-> Natural 20 -> Natural 20 -> Meet or Exceed AC = Insta-Kill

conversely

-> Miss AC = Miss
-> Natural 1 -> Meet or Exceed AC = 50% damage
-> Natural 1 -> Miss AC = Miss
-> Natural 1 -> Natural 1 -> Meet or Exceed AC = Broken weapon, no one is hurt
-> Natural 1 -> Natural 1 -> Miss AC = Broken weapon, shards hurt you or allies

[Gwendol]

I don't find the miss rule to be so bad. It introduces an element of unpredictability that sometimes is needed to keep an edge to the game.

[truemane]

The major problem with critical misses, as has been noted many times, is that it penalizes melee characters without offering an equal penalty to spellcasters. And melee characters have it rough enough as it is.

[Gwendol]

True. Roll a d20, spell fizzles on a 1. Fixed!

[Snowbluff]

The major problem with critical misses, as has been noted many times, is that it penalizes melee characters without offering an equal penalty to spellcasters. And melee characters have it rough enough as it is.

I cast Dominate!
DM: The creature rolls a nat 20. Fail. (5%)

I cast Orb of Ice
DM: You rolled a nat 1. Fail. (5%)

The only place critical failures and successes don't apply are a few spells. The penalty isn't equal, but you're implying an insurmountable gap.

[prufock]

I'm not talking about any critical fumble rules here, I'm talking about the core rule of always missing an attack or failing a save on a natural roll of 1. I can see the symmetry here: 20 always hits, 1 always misses. But is that even a good argument for that rule?
It eliminates guaranteed success in combat. Otherwise the attacking player would not need to roll at all, except to determine if it was a regular hit or a critical. Critical hits would not need to be confirmed.


I mean, regardless of training, you can't shake off that 5% chance of missing the attack or failing that save.
Crap happens. Your blade glances off a plate of armor. The enemy moves a certain way at just the right time. The sweat on your handle causes your axe to slip just a little. From a realism perspective, a flat percentage doesn't make sense, but even great marksmen can have a bad day, or suffer environmental interference, or simply screw up (rarely).


What's worse, that rule penalizes harder the characters that roll more dice, which means the characters that have higher BAB, generally. If the character is getting better at hitting something, why does he have a higher statistical chance of automatically missing than someone who's worse at hitting something?
He doesn't. Each attack has a 5% chance of missing, period. Full BAB classes just get 33.3% more attacks than 3/4 BAB classes and 100% more than 1/2 BAB classes.

Fighter 20: BAB 20/15/10/5, hits 76 times over 20 full attacks, misses 4.
Rogue 20: BAB 15/10/5, hits 57 times over 20 full attacks, misses 3.
Wizard 20: BAB 10/5, hits 38 times over 20 full attacks, misses 2.

This is ignoring any discrepancy due to the higher attack bonuses as well. The fighter, if built strongly, is probably ONLY missing on a 1, while the others will have a larger probability of missing due to their lower BAB.

[Mighty_Chicken]

I think that automatic failures are ok for combat. Even the best warrior make mistakes, and in the end he'll still defeat those who are weaker.

The problem is when a single roll defines your success in an activity. For example, skills. In my skill system (http://www.giantitp.com/forums/showthread.php?t=321039) characters who spent enough skill points in a skill don't automatically fail with a 1. The best lock picker in the multiverse shouldn't have a 5% chance to fail at lock picking a rusty, mundane lock.

As for critical failures... they're awesome! Even for skills! But they should be optional - then again, both people who are good and bad at something can do something amazing (a natural 20), but people who are good at something should almost never so something amazingly bad.

So as a house rule in the campaign I'm about to start, if you roll a natural 1, I challenge you (as per FATE) to accept a critical failure. If you do, I, the DM, give you an action point token, and you screw up in an embarrassing anh humiliating way, which your fellow player can help me to describe. If you don't, it's a normal natural 1 roll (an automatic failure, except with skills as mentioned above).

[subject42]

The major problem with critical misses, as has been noted many times, is that it penalizes melee characters without offering an equal penalty to spellcasters. And melee characters have it rough enough as it is.

You can actually ameliorate this somewhat. It's not perfect, but if you look at something like Paizo's crit fail deck, they break it up into ranged, natural/unarmed, melee, and magic categories.

If you crit fail in a type where the most common mechanic is to make a lot of attacks (eg: ranged or unarmed), the result is pretty mild.

If you crit fail in a type where failure is uncommon, like magic, you're in for a world of hurt.

[zlefin]

I think that automatic failures are ok for combat. Even the best warrior make mistakes, and in the end he'll still defeat those who are weaker.

The problem is when a single roll defines your success in an activity. For example, skills. In my skill system (http://www.giantitp.com/forums/showthread.php?t=321039) characters who spent enough skill points in a skill don't automatically fail with a 1. The best lock picker in the multiverse shouldn't have a 5% chance to fail at lock picking a rusty, mundane lock.

As for critical failures... they're awesome! Even for skills! But they should be optional - then again, both people who are good and bad at something can do something amazing (a natural 20), but people who are good at something should almost never so something amazingly bad.

So as a house rule in the campaign I'm about to start, if you roll a natural 1, I challenge you (as per FATE) to accept a critical failure. If you do, I, the DM, give you an action point token, and you screw up in an embarrassing anh humiliating way, which your fellow player can help me to describe. If you don't, it's a normal natural 1 roll (an automatic failure, except with skills as mentioned above).

it's already the case by RAW that a natural 1 is not an automatic failure on a skill roll.

[Rebel7284]

The problem is when a single roll defines your success in an activity. For example, skills. In my skill system (http://www.giantitp.com/forums/showthread.php?t=321039) characters who spent enough skill points in a skill don't automatically fail with a 1. The best lock picker in the multiverse shouldn't have a 5% chance to fail at lock picking a rusty, mundane lock.

Skills don't fail on a natural 1 anyway... this rule only applies to attacks and saves... this is why Moment of Perfect Mind is so good...

[Kraken]

You should probably remove the word critical from the title, as it implies something extra happens when you fail a save.

I think with attack rolls missing on a 1 is fine, but I'm not so sure about automatic failure for saves. I don't have a reason for this, just a gut feeling.

[Larkas]

My 2 cp:

In my games, a Nat 20 is a crit threat. Afer that, you must roll again. Hitting the target AC is a confirmed crit, not doing so is still a regular hit. Another 20 is a insta-kill threat, after which you roll a third time, and the same rules apply.

So, for a Nat 1, call it a fail threat. After that, roll again. Hitting the target AC, you're ok; call it a "near miss," and damage is halved. Miss the target AC on the second roll, (but don't roll a 1), and you've got a miss on your hands. Two 1s and another fail, you're into broken-weapon/stabbed your friend territory.

To illustrate: (with -> indicating a die roll)
-> Meet or exceed AC = Hit
-> Natural 20 -> Miss AC = Hit
-> Natural 20 -> Meet or Exceed AC = Critical Hit
-> Natural 20 -> Natural 20 -> Miss AC = Critical Hit
-> Natural 20 -> Natural 20 -> Meet or Exceed AC = Insta-Kill

conversely

-> Miss AC = Miss
-> Natural 1 -> Meet or Exceed AC = 50% damage
-> Natural 1 -> Miss AC = Miss
-> Natural 1 -> Natural 1 -> Meet or Exceed AC = Broken weapon, no one is hurt
-> Natural 1 -> Natural 1 -> Miss AC = Broken weapon, shards hurt you or allies

I don't care at all for critical fumble and insta-kill house rules, but aside from that, that's not a bad system.


I don't find the miss rule to be so bad. It introduces an element of unpredictability that sometimes is needed to keep an edge to the game.

I understand what you're saying, but aren't challenging opponents for that?


The major problem with critical misses, as has been noted many times, is that it penalizes melee characters without offering an equal penalty to spellcasters. And melee characters have it rough enough as it is.

Eh, you can score a critical success in a save against a spell, or you can critically miss a touch/ranged touch spell. That doesn't mean it's an equal penalty, of course. Casters have access to spells that automatically hit and don't offer saves, and some spells damage even on a successful save, but that's beyond the point.


True. Roll a d20, spell fizzles on a 1. Fixed!

Eh, that works, but I'm not sure it makes the system better.


I cast Dominate!
DM: The creature rolls a nat 20. Fail. (5%)

I cast Orb of Ice
DM: You rolled a nat 1. Fail. (5%)

The only place critical failures and successes don't apply are a few spells. The penalty isn't equal, but you're implying an insurmountable gap.

I don't think that was directed to me, but just in case: I'm not implying nothing of the sort, ok? :smallsmile:


It eliminates guaranteed success in combat. Otherwise the attacking player would not need to roll at all, except to determine if it was a regular hit or a critical. Critical hits would not need to be confirmed.

I understand that. I'm not so sure that's a bad thing.


Crap happens. Your blade glances off a plate of armor. The enemy moves a certain way at just the right time. The sweat on your handle causes your axe to slip just a little. From a realism perspective, a flat percentage doesn't make sense, but even great marksmen can have a bad day, or suffer environmental interference, or simply screw up (rarely).

I also understand that. A flat percentage might even make sense, just a much lower percentage.


He doesn't. Each attack has a 5% chance of missing, period. Full BAB classes just get 33.3% more attacks than 3/4 BAB classes and 100% more than 1/2 BAB classes.

Fighter 20: BAB 20/15/10/5, hits 76 times over 20 full attacks, misses 4.
Rogue 20: BAB 15/10/5, hits 57 times over 20 full attacks, misses 3.
Wizard 20: BAB 10/5, hits 38 times over 20 full attacks, misses 2.

This is ignoring any discrepancy due to the higher attack bonuses as well. The fighter, if built strongly, is probably ONLY missing on a 1, while the others will have a larger probability of missing due to their lower BAB.

Okay, sorry, I could've been clearer. A character with a higher BAB has a higher statistical chance of automatically missing at least once during a full attack than a character with a lower BAB. I'm purposefully ignoring the discrepancy due to higher attack bonus, because it doesn't add much to the issue at hand. I don't have a problem with regular misses, my problem is with critical misses. (Let's not forget: for a poorly built high BAB, high level character, half his attacks are as good as those of a poor BAB, high level character. This doesn't have much bearing on this discussion, though.)


I think that automatic failures are ok for combat. Even the best warrior make mistakes, and in the end he'll still defeat those who are weaker.

The problem is when a single roll defines your success in an activity. For example, skills. In my skill system (http://www.giantitp.com/forums/showthread.php?t=321039) characters who spent enough skill points in a skill don't automatically fail with a 1. The best lock picker in the multiverse shouldn't have a 5% chance to fail at lock picking a rusty, mundane lock.

As for critical failures... they're awesome! Even for skills! But they should be optional - then again, both people who are good and bad at something can do something amazing (a natural 20), but people who are good at something should almost never so something amazingly bad.

So as a house rule in the campaign I'm about to start, if you roll a natural 1, I challenge you (as per FATE) to accept a critical failure. If you do, I, the DM, give you an action point token, and you screw up in an embarrassing anh humiliating way, which your fellow player can help me to describe. If you don't, it's a normal natural 1 roll (an automatic failure, except with skills as mentioned above).

A common misconception. Natural 1 on skills are never an automatic failure under the stock rules. Regardless, like I said above, I don't care at all about critical fumble rules, though yours appears to be less aggravating than most.


You should probably remove the word critical from the title, as it implies something extra happens when you fail a save.

I think with attack rolls missing on a 1 is fine, but I'm not so sure about automatic failure for saves. I don't have a reason for this, just a gut feeling.

But it does! You fail when you really shouldn't have! :smallbiggrin: I must confess I don't know why people adopt critical fumble rules for attacks but not for saves. Ease of abstraction, I guess?

[Ydaer Ca Noit]

Are you talking about missing on 1, or missing on 1 and something bad happens?

In my opinion missing on a 1 is ok rule. As is hitting on a 20.

Extra effects like hitting your ally or chopping hands off randomly is something else that I hate (because no one cares if one NPC loses a hand, but its a different story for a PC. Its not ok for it to happen randomly, for me)

[Larkas]

Are you talking about missing on 1, or missing on 1 and something bad happens?

In my opinion missing on a 1 is ok rule. As is hitting on a 20.

Extra effects like hitting your ally or chopping hands off randomly is something else that I hate (because no one cares if one NPC loses a hand, but its a different story for a PC. Its not ok for it to happen randomly, for me)

Read the first sentence in the original post. Your opinion is welcome, but take the time to at least read the first post in the thread. :smallwink:

[zephyrkinetic]

I don't care at all for critical fumble and insta-kill house rules, but aside from that, that's not a bad system.

I did oversimplify a bit, in an effort to keep things general. Insta-kills don't apply on anything above an average CR. Fumbles apply to magic weapons; breaks only apply to mundane. And I think anyone that rolls 2x20s or 2x1s and then confirms the hit/miss has done something extraordinary, whether it be good or bad.

[Suddo]

True. Roll a d20, spell fizzles on a 1. Fixed!

So spell resist... I mean that is why I like having spell resistance on my character it allows spells to fumble.


Anyways back to the topic at hand. I like it for the same reason I like critical hits. It makes it so the die has to be rolled, there is never an auto-succeed. It should be noted that although odd I dislike critical (hit or fumbles) on skills, though that is due to different reasons.

[Defiled Cross]

From both an Adventurer and Dungeon Master perspective, I've always thought the Critical Miss rule to be a just one.

Besides..

..rolling your ankle in a Mole-hole while swinging at a injured, low-level Goblin is always a good laugh.

:smallamused:

[prufock]

Okay, sorry, I could've been clearer. A character with a higher BAB has a higher statistical chance of automatically missing at least once during a full attack than a character with a lower BAB.
Well yes, but that's just by way of more attacks. They also have a higher chance of hitting at least once during a full attack than a character with a lower BAB. It's the higher number of hits that matter, since that's the source of damage.

Part of your objection might be from conceptualizing it. Both a +10/+5 BAB and a +20/+15/+10/+5 BAB character are making multiple attempts to strike during a full attack. A full attack isn't just swing-swing-swing-swing, it's swing-position-parry-swing-thrust-block-duck-swing-swing-slash-stab-deflect-swing for both characters. The higher BAB just represents a greater chance of those swings and thrusts and stabs and slashes actually connecting.

If it still bothers you, use the following variant from the DMG: A roll of 1 is treated as -10, a roll of 20 is treated as 30. Critical threats are unaffected, but now a high BAB character can still hit on a 1, and a low BAB character can still miss on a 20.

[Larkas]

Well yes, but that's just by way of more attacks. They also have a higher chance of hitting at least once during a full attack than a character with a lower BAB. It's the higher number of hits that matter, since that's the source of damage.

Hmmm, makes sense. But I don't know, something about it still bothers me... You see, the attacks after the first and maybe the second are hard enough to hit, I don't think the "primary" attacks need a 5% chance to miss automatically.


Part of your objection might be from conceptualizing it. Both a +10/+5 BAB and a +20/+15/+10/+5 BAB character are making multiple attempts to strike during a full attack. A full attack isn't just swing-swing-swing-swing, it's swing-position-parry-swing-thrust-block-duck-swing-swing-slash-stab-deflect-swing for both characters. The higher BAB just represents a greater chance of those swings and thrusts and stabs and slashes actually connecting.

That wasn't part of the objection, but I must say I never wondered how that would translate to what "we would see". That's a nice imagery! Thanks a lot for that, it was very instructive! :smallsmile:


If it still bothers you, use the following variant from the DMG: A roll of 1 is treated as -10, a roll of 20 is treated as 30. Critical threats are unaffected, but now a high BAB character can still hit on a 1, and a low BAB character can still miss on a 20.

Hey, I like that idea! And I must've completely skimmed over it, since I don't remember reading about it! Care to share where exactly can I find it?

[Ydaer Ca Noit]

Read the first sentence in the original post. Your opinion is welcome, but take the time to at least read the first post in the thread. :smallwink:

Well, other guys said about fumbles, and I wanted to say my opinion about them too. I apologize that a strayed so far from your original question, I didn't think that it would cause trouble.

[eggynack]

Well, other guys said about fumbles, and I wanted to say my opinion about them too. I apologize that a strayed so far from your original question, I didn't think that it would cause trouble.
Well, you have. Whole worlds of trouble, bourne of your comment. It shall consume the forum in a great hellfire, I think.

Anyway, I agree. We could play the game where we try to invent incredibly complicated rules that somehow hit magic and mundane characters equally, because it's a really difficult thing (even if you just assign a new roll to all spells, wizards have less to fear of failure, as they rely less on in combat actions), or we could just ditch the rule. The latter seems better. It doesn't make much sense that a master swordsman would literally fail to hit the broad side of a barn one in twenty times, and it's especially problematic because that's all the rule is really causing. If your enemy is actually hard to hit, then that can be represented with higher AC, and automatic failure will only come up against chubby wizards and random rubes.

[Eldonauran]

I don't have an issue with the auto hit/miss or the auto pass/fail system. Even if you break it down into "I'm an expert and there is no way I'm going to miss 5% of the time" argument. The other party you are engaging in combat (whether expert swordsman or a monster) is either just as good as you are, or nearly so. They get the same mechanical benefits and drawbacks as the players in this regard.

Aside from that, taking away auto hit/miss/pass/fail might be an option for your game. Be sure to apply that rule to all aspects equally. You roll a 20 on a save? Confirm it.

If you are fighting opponents that are not close to your CR, they are obviously not challenging and won't apply to this. Why engage them anyway?

[purpenflurb]

As someone who has been a martial arts student/teacher for over a decade, the rule is actually pretty reasonable. Even the best trained fighters make mistakes. Whether it be targeting your blow completely ineffectively, or actually hurting yourself, it is a rather realistic rule.

For a more obvious comparison, look at the olympics. If you have been watching, on more than one occasion, extremely good athletes just messed up. Trying to perform at the edge of their game they lost control completely. Given as these are theoretically the best at their given disciplines, that should be a pretty good illustration of why that failure chance is realistic.

[eggynack]

For a more obvious comparison, look at the olympics. If you have been watching, on more than one occasion, extremely good athletes just messed up. Trying to perform at the edge of their game they lost control completely. Given as these are theoretically the best at their given disciplines, that should be a pretty good illustration of why that failure chance is realistic.
I think there are a few questions that must be asked about such a comparison. First, how often did they screw up in that manner? Was in one in twenty, less, or more? Second, should their failure truly be qualified as one defined by a roll of one on their rhetorical dice, or were the things they were trying to do difficult to the level that a failure would fall under an ordinary bad roll? Third, as a continuation of the second question, would they still have failed that amount of the time were they faced with a less difficult challenge? Would a fencer, for example, still miss completely were they faced with me, a crappy non-fencer, instead of their Olympic level opponent? Finally, are these people truly the theoretical best at their given discipline? In other words, while they might be the best in our world, would they still be the best in D&D world, where god-like beings roam the earth? These seem like rather important things, I think.

[Larkas]

Well, other guys said about fumbles, and I wanted to say my opinion about them too. I apologize that a strayed so far from your original question, I didn't think that it would cause trouble.

No harm done. :smallsmile:


Well, you have. Whole worlds of trouble, bourne of your comment. It shall consume the forum in a great hellfire, I think.

Anyway, I agree. We could play the game where we try to invent incredibly complicated rules that somehow hit magic and mundane characters equally, because it's a really difficult thing (even if you just assign a new roll to all spells, wizards have less to fear of failure, as they rely less on in combat actions), or we could just ditch the rule. The latter seems better. It doesn't make much sense that a master swordsman would literally fail to hit the broad side of a barn one in twenty times, and it's especially problematic because that's all the rule is really causing. If your enemy is actually hard to hit, then that can be represented with higher AC, and automatic failure will only come up against chubby wizards and random rubes.

Exactly! I mean, I could see some chance of mistake, but certainly not 1-in-20.


I don't have an issue with the auto hit/miss or the auto pass/fail system. Even if you break it down into "I'm an expert and there is no way I'm going to miss 5% of the time" argument. The other party you are engaging in combat (whether expert swordsman or a monster) is either just as good as you are, or nearly so. They get the same mechanical benefits and drawbacks as the players in this regard.

Not necessarily. You could be a level 20 two-weapon fighting ranger swimming in a sea of level 1 warriors, with bonuses to attack through the roof and no chance of defeat whatsoever: you'd still miss 5% of your attacks. Heck, like eggynack said, you could be attacking a wall and you'd still miss 5% of the time. Let's remember we're talking about a character who's at the pinnacle of swordsmanship here, a level of swordplay no real human could hope to achieve, and he's got magical assistance to boot. He still has 1-in-20 chance to screw it up.

Unless you meant to say that the rule will affect equally both sides of the conflict. That's pretty obvious, since the rule is the same for both. But a monster not connecting a sure hit isn't anticlimatic; a PC doing so might be. Changing the rule, however, would also affect both sides equally. I don't really see an issue here.


Aside from that, taking away auto hit/miss/pass/fail might be an option for your game. Be sure to apply that rule to all aspects equally. You roll a 20 on a save? Confirm it.

I don't intend to remove auto hit/pass. Like I said, I don't care at all for this symmetry.


If you are fighting opponents that are not close to your CR, they are obviously not challenging and won't apply to this. Why engage them anyway?

Because they might want to engage you? I don't know. :smallconfused:


As someone who has been a martial arts student/teacher for over a decade, the rule is actually pretty reasonable. Even the best trained fighters make mistakes. Whether it be targeting your blow completely ineffectively, or actually hurting yourself, it is a rather realistic rule.

For a more obvious comparison, look at the olympics. If you have been watching, on more than one occasion, extremely good athletes just messed up. Trying to perform at the edge of their game they lost control completely. Given as these are theoretically the best at their given disciplines, that should be a pretty good illustration of why that failure chance is realistic.


I think there are a few questions that must be asked about such a comparison. First, how often did they screw up in that manner? Was in one in twenty, less, or more? Second, should their failure truly be qualified as one defined by a roll of one on their rhetorical dice, or were the things they were trying to do difficult to the level that a failure would fall under an ordinary bad roll? Third, as a continuation of the second question, would they still have failed that amount of the time were they faced with a less difficult challenge? Would a fencer, for example, still miss completely were they faced with me, a crappy non-fencer, instead of their Olympic level opponent? Finally, are these people truly the theoretical best at their given discipline? In other words, while they might be the best in our world, would they still be the best in D&D world, where god-like beings roam the earth? These seem like rather important things, I think.

Pretty much what eggynack said! Thanks a lot for the input, purpenflurb, but think about this for a second: how many times have you seen a black belt missing a serious (i.e.: not training) blow against a fresh white belt? How many times have you seen a pole vaulting athlete, accustomed to jump almost 6m, missing a 2m jump? This is what "missing on a 1" would look like. The white belt doesn't have the "deflection AC" a fellow black belt could have had. He's literally a sitting duck against a black belt. We regularly see athletes messing up during the Olympics because they're operating at the top of their capacity there. Hell, if an "olympic encounter" could be quantified in CR, I could see it very well in the "very difficult" terrain, i.e.: CR+5. But the same athletes wouldn't mess up quite as frequently if doing "leisurely athletics", something that could be almost impossible for us non-athletic types, but is a walk in the park for them. I'm not saying they wouldn't mess up at all, I'm just saying that 5%, or 1 in 20 punches, or 1 in 20 jumps, is way too much.

[Zweisteine]

There is some sense in automatic misses, as demonstrates by the fact that even the best swordsman sometimes fumbles. The problem with the rule, then, is that the best swordsman in reality stands nowhere close to the best swordsman in D&D (and by best swordsman, I mean level 20 fighter).
As so many have stated above, even olympians make mistakes. However, most of reality maps to the lower levels of a D&D character (even if there isn't an effective maximum level of five in real life, we certainly don't get to level 10, much less 20).
Here's the houserule I just now made up to fix this: If you roll a natural one on an attack roll (but still would have hit, which really shouldn't be happening), roll again. If you roll under your base attack bonus, you do not automatically miss (but you still rolled a one). This can be applied to saves as well, but roll a d12 for the failure chance as well (though this lessens the benefits of certain abilities that prevent autofailing on saves).

There's also always this simple alternative: Treat a natural 20 as a roll of 40, and a natural 1 as a -20 (or a -19, if you really love your equality, I suppose). That prevents commoners from hitting 20th level characters, and prevents those characters from missing commoners.

[Chronos]

Wait, if I understand you correctly, you're fine with natural 20s auto-succeeding on both saves and attacks, but not with natural 1s auto-failing on those? That seems like an inconsistent position to take, to me, since a success on an attack roll is what the attacker wants, but a success on a save is what the defender wants. Surely, it would make more sense, if we're going to be removing half the cases, to either remove the two that favor defenders, or to remove the two that favor attackers.

(Actually, I can see one rationale, in that mundanes usually do things with attack rolls and spellcasters usually do things with saves, and so using nat 20s but not nat 1s favors the mundanes a bit. But I don't think that's what the OP is getting at.)

[Dimers]

You should probably remove the word critical from the title, as it implies something extra happens when you fail a save.

Actually, something extra does happen -- your items have to make saves if you rolled a 1.

[eggynack]

Wait, if I understand you correctly, you're fine with natural 20s auto-succeeding on both saves and attacks, but not with natural 1s auto-failing on those? That seems like an inconsistent position to take, to me, since a success on an attack roll is what the attacker wants, but a success on a save is what the defender wants. Surely, it would make more sense, if we're going to be removing half the cases, to either remove the two that favor defenders, or to remove the two that favor attackers.

(Actually, I can see one rationale, in that mundanes usually do things with attack rolls and spellcasters usually do things with saves, and so using nat 20s but not nat 1s favors the mundanes a bit. But I don't think that's what the OP is getting at.)
That is a good point, or perhaps points. Auto-whatevers are kinda weird, but just as I am hesitant to make any change to the system that favors casters, I am equally hesitant to make any change that hurts mundanes. Perhaps it can be considered a neutral change if you remove both aspects of the system, but I'm not sure.

[Rejusu]

The question is does it really matter? If you're rolling a one and have a big enough bonus to still hit their AC then you're certainly going to land every other attack you make. If you roll a one and your bonus isn't enough to hit their AC then the fact it's an automatic miss is irrelevant. Frankly it's pretty much a non issue, it's only fumble rules I take umbrage with.

"You roll a 1 and stab yourself in the foot" :smallmad:

That's silly. But as for the miss rules? Seven to eight times out of ten if you roll a one you wouldn't be hitting anyway. And if your to hit is that high or their AC is that low you're guaranteed to land the rest of your attacks so one missed attack isn't the end of the world.

[purpenflurb]

I think there are a few questions that must be asked about such a comparison. First, how often did they screw up in that manner? Was in one in twenty, less, or more? Second, should their failure truly be qualified as one defined by a roll of one on their rhetorical dice, or were the things they were trying to do difficult to the level that a failure would fall under an ordinary bad roll? Third, as a continuation of the second question, would they still have failed that amount of the time were they faced with a less difficult challenge? Would a fencer, for example, still miss completely were they faced with me, a crappy non-fencer, instead of their Olympic level opponent? Finally, are these people truly the theoretical best at their given discipline? In other words, while they might be the best in our world, would they still be the best in D&D world, where god-like beings roam the earth? These seem like rather important things, I think.

To address your points... 1/20 seems a ballpark number, from what I have seen, though some sports seem to be higher. And some of them are clearly "critical" failures, if you will. In terms of fighting, even faced with a less difficult challenge, failure is certainly always a possibility. I have seen more than my share of people hurt themselves while trying to attack. Or lose control of their momentum, or hit the wrong thing, etc. It may not fit "Exactly", but even for nearly superhumans I don't think that sort of failure rate is totally unreasonable. Especially when paired with an also possibly unrealistic critical rate.

[Eldonauran]

Not necessarily. You could be a level 20 two-weapon fighting ranger swimming in a sea of level 1 warriors, with bonuses to attack through the roof and no chance of defeat whatsoever: you'd still miss 5% of your attacks. Heck, like eggynack said, you could be attacking a wall and you'd still miss 5% of the time. Let's remember we're talking about a character who's at the pinnacle of swordsmanship here, a level of swordplay no real human could hope to achieve, and he's got magical assistance to boot. He still has 1-in-20 chance to screw it up.

Unless you meant to say that the rule will affect equally both sides of the conflict. That's pretty obvious, since the rule is the same for both. But a monster not connecting a sure hit isn't anticlimatic; a PC doing so might be. Changing the rule, however, would also affect both sides equally. I don't really see an issue here.


Well, let me add to my comments before. I believe the auto hit/miss/pass/fail rules were intended to reflect equally powerful creatures (or near so) engaging in combat with each other. If you are fighting things well over (or under) the CR, there is going to be major discrepancies in how combat turns out. The table given by the DMG showing exp per CR gives weight to this as you shouldn't be engaging things of certain CR so far removed relative to your level.

That said, I completely disregard the example of a level 20 warrior fighting a horde of lower level creatures. Sheer weight of numbers should overwhelm a powerful creature but isn't represented in the D&D rules.

If you want to house rule a lesser chance of auto missing/failing, I say go for it. At least be consistant with the auto hit/pass rules as well. Not every fight is going to have a climax and I've had many fights that are full of anti-climax moments (like missing 5 turns in a row, the whole party and the enemies too). There is always an element of chance in the D&D (we use dice after all).

[Fiery Diamond]

Wait, if I understand you correctly, you're fine with natural 20s auto-succeeding on both saves and attacks, but not with natural 1s auto-failing on those? That seems like an inconsistent position to take, to me, since a success on an attack roll is what the attacker wants, but a success on a save is what the defender wants. Surely, it would make more sense, if we're going to be removing half the cases, to either remove the two that favor defenders, or to remove the two that favor attackers.

(Actually, I can see one rationale, in that mundanes usually do things with attack rolls and spellcasters usually do things with saves, and so using nat 20s but not nat 1s favors the mundanes a bit. But I don't think that's what the OP is getting at.)

Look at it this way: The one who is being tested and tried is the one whom the rules make roll - the game doesn't measure your defensive (AC) abilities as an actual activity, just a static thing that the activity (Attacking) uses as a comparison to determine success or failure; on the other hand, for reactive/resisting (Saves) abilities, the reaction/resistance is the activity, and the offensive threat is static. Removing the 1s rule but not the 20s says: activities which are variable do not have a guaranteed minimum failure chance, but good luck can happen and so do have a guaranteed minimum success chance (reduces negative-outcome variance for the players, since all variance mostly hinders players; but it does allow for moment of awesome).

I'd rather break from the default in a different way: "Characters" keep auto-successes and lose auto-failures; "Non-characters" lose auto-successes and keep auto-failures. The terms in quotes are metagame concepts; basically, anyone important enough that their unexpected success can be seen as interesting, rather than merely a #$%^y die roll, should keep auto-success rules and lose auto-failures. This would include the PCs and most named characters, presumably. Mooks, minions, legions of kobolds/undead/animals/etc. would keep auto-failures to give the "characters" a chance of unexpected success, but lose auto-successes to prevent a "character" from suffering from unfortunate die rolls made by the "non-character."


In short:

Characters (PCs, named characters, bosses, etc.)
-1s rule (nope)
-20s rule (yep)

Non-characters [NCs] (minions, mooks, hordes, random baddies, etc.)
-1s rule (yep)
-20s rule (nope)

You could even take it further than that if you wanted and apply the same logic to the existence of critical hits, removing or reducing critical chances for non-characters while leaving critical rules intact for characters. Something like:

-NCs have a critical threat only on natural 20s, regardless of their critical range. In order to confirm a critical, they must not only hit AC on confirmation roll but also have their natural die result on the confirmation roll be in their critical range.

Example: Mook A is using a rapier (18-20/x2).
Miss AC = Miss
Hit AC = Hit
Nat 20 -->
a) This roll missed AC? MISS (NC has no chance of hitting)
b) This roll hit AC? Critical threat. Go to next step.
Nat 20 --> Roll other than Nat 18, 19, or 20 = Hit
Nat 20 --> Nat 18, 19, or 20 = Critical Hit

After all, we're not talking about realistic-ness, we're talking about what makes the game fun.

[prufock]

Hey, I like that idea! And I must've completely skimmed over it, since I don't remember reading about it! Care to share where exactly can I find it?

Sidebar on p. 25 of DMG.

[Nibbens]

In my games, a Nat 20 is a crit threat. Afer that, you must roll again. Hitting the target AC is a confirmed crit, not doing so is still a regular hit. Another 20 is a insta-kill threat, after which you roll a third time, and the same rules apply.

So, for a Nat 1, call it a fail threat. After that, roll again. Hitting the target AC, you're ok; call it a "near miss," and damage is halved. Miss the target AC on the second roll, (but don't roll a 1), and you've got a miss on your hands. Two 1s and another fail, you're into broken-weapon/stabbed your friend territory.

Stealing this for my games! lol. I've been looking for a new way to spice up crit fails and crit hits.

[zephyrkinetic]

Stealing this for my games! lol. I've been looking for a new way to spice up crit fails and crit hits.

Glad I could help. Let me know how it works out for you and your table.

[Scow2]

A natural 1, while it may actually occur 5% of the time, is actually a one-in-a-million chance - but the rules of D&D say there's a 5% chance of the universe kicking probability to the curb allowing a longshot or thwarting a sure thing.

Or, you can take a Nat1 on an attack roll to be the equivalent of the defender rolling a Nat 20 on a defense roll. We're not talking about someone trying to achieve a static result against their own ability like jumping a gap or over an obstacle - we're talking about someone taking action against another living (Or unliving), active agent who is doing everything in their power to keep themselves alive and unharmed, no matter how outclassed they might be.