D&D All. What's your take on Critical Faliures?

[Jon_Dahl]

You're not familiar with gamers OR the Sunk Cost Fallacy, are you?

No, but thank you for making me read up on the theory. That was interesting.

[Prince Raven]

You make a valid point, but that's not the case here. For two years or so, one of the players travelled 400 km in total per a game session. There would have been games closer by.
(I know that it seems that I'm making this up, but I'm not)

What I'm saying is "my players like them, therefore they're good" is just as valid as saying "my players hate them, therefore they're bad". It's a very small sample base you're working with.

[NichG]

So don't be boring, don't just say "you miss the duck", put a little bit of roleplaying into your combat. You don't have to fix everything by adding in unnecessary mechanics. I say unnecessary because, as you just stated, there already is a failure mechanic built into the base rules.

The reason this is boring, and that narrating things doesn't fix the underlying problem, comes from the fact that fighting in D&D is basically an exercise in watching the central limit theorem be generated. E.g. you're rolling multiple times to answer a very simple question at the end of the day: 'how many attacks does it take to get to the center of this orc?' Its such that with a little mathematics, you could compress the entirety of a fighter's participation in combat to a single d100 roll and a chart lookup telling you how many rounds the fighter needs to keep attacking the enemy until it dies.

Part of this is that if you're just talking about hits or misses, the eventual consequence is totally independent of context. Lets say it takes 8 hits to kill the dragon. The fighter makes 16 attacks and the dragon dies. This could happen because the fighter missed 8 times in a row and then hit 8 times in a row, or interchanged miss/hit/miss/hit/..., or had some other pattern. But at the end of the day, all of that context is removed when you just turn it into adjustments in a hitpoint track. The only time the context matters is if you can kill the dragon with fewer attacks, e.g. it has to do with the positioning of the last of the hits (but the rest don't matter). This also means that usually you never have to change your plans on the basis of how well the fighter does on their attack sequence - just wait it out and the central limit theorem will make everything nice and reliable for you.

That's why even variations of 'you miss the duck' aren't going to be all that interesting.

If you want to make those individual die rolls more meaningful you have to make them non-commutative somehow. Critical failure charts are one way to do this, because the consequence of a critical failure early on in the fight is going to be different than the consequence of a critical failure near the end of the fight. Early on, a debuff is more harmful. Later on, additional damage to the attacker is more harmful. Either case may involve a change in the plan of action to accommodate its influence.

But in general it comes down to there being more interesting questions one can ask using random generators than 'how long will it take me to succeed?' or 'did I succeed or fail?'. Dice systems where specific dice outcomes mean something in addition to the thing being tested help with this, for example, but it requires a bit of caution to make sure that they don't end up being really metagamey.

[Prince Raven]

Well, maybe my players and I just haven't played enough to get that jaded.

[Jon_Dahl]

Well, maybe my players and I just haven't played enough to get that jaded.

What I'm saying is "my players aren't that jaded, therefore it's still interesting" is just as valid as saying "my players are jaded, therefore it isn't that interesting". It's a very small sample base you're working with.

[Prince Raven]

What I'm saying is "my players aren't that jaded, therefore it's still interesting" is just as valid as saying "my players are jaded, therefore it isn't that interesting". It's a very small sample base you're working with.

Spoiler

Show

In all seriousness though, I really don't understand this whole "combat encounters are boring" thing you're talking about, for a lot of players combat is the most exciting part. D&D is a combat-focussed game, after all.

[JaminDM]

According to this rule, when a fighter and a cleric of 6th level with identical equipment, Dex and combat buffs fight against identical enemies, the Fighter is more likely to get hit. The Fighter attacks twice, the Cleric only once. That's because for each attack, there's a 2.5% chance of drawing attacks of opportunity.


That's kinda stupid, don't you think? Being a better fighter means you make more mistakes.

Good point...

Okay... I think this can be fixed by adding your BAB+Dex modifier to the confirmation roll that has to be less than 20 to confirm a critical failure.

[NichG]

Spoiler

Show

In all seriousness though, I really don't understand this whole "combat encounters are boring" thing you're talking about, for a lot of players combat is the most exciting part. D&D is a combat-focussed game, after all.

Combat is interesting, but it has its ups and downs. The ups are hinge points where the future of the fight is in flux because of something or other. The downs are sequences where you're waiting to do the detailed evaluation of what is basically a foregone conclusion (4ed seems to in particular have a problem with this kind of 'down'). Its very exciting when some spell hits the field and now I have to evaluate whether to keep the same target or move out of a damaging area or seek healing or whatever. Its exciting when people are rolling saving throws against things which, if they fail, will really change their available options (now you're blind - what do you do?). But its pretty boring when you're waiting for someone to roll through his full attack: its boring because you don't really need to pay attention, but at the same time its something that has to get done for the combat to move on.

This is more of a problem with full attacks being somewhat poorly designed mechanics than anything else in particular. My personal preferred fix for that is homebrew and martial options that reduce the need for people to make use of long attack sequences (e.g. stuff that lets you sacrifice attacks for fixed buffs to a single attack, maneuvers that resolve as a single attack, etc) but critical successes/failures with 'special effects' can serve that goal as well.

[Alex12]

The only times I can see critical fumbles being reasonable and not kind of dumb in the games I play is if it was only applied to some characters. Say, half-BAB characters, no relevant feats or class abilities, below-average Str or Dex (or Wis or whatever you're using to determine to-hit) as appropriate, and no magic weapon.

[Kelb_Panthera]

Combat is interesting, but it has its ups and downs. The ups are hinge points where the future of the fight is in flux because of something or other. The downs are sequences where you're waiting to do the detailed evaluation of what is basically a foregone conclusion (4ed seems to in particular have a problem with this kind of 'down'). Its very exciting when some spell hits the field and now I have to evaluate whether to keep the same target or move out of a damaging area or seek healing or whatever. Its exciting when people are rolling saving throws against things which, if they fail, will really change their available options (now you're blind - what do you do?). But its pretty boring when you're waiting for someone to roll through his full attack: its boring because you don't really need to pay attention, but at the same time its something that has to get done for the combat to move on.

This is more of a problem with full attacks being somewhat poorly designed mechanics than anything else in particular. My personal preferred fix for that is homebrew and martial options that reduce the need for people to make use of long attack sequences (e.g. stuff that lets you sacrifice attacks for fixed buffs to a single attack, maneuvers that resolve as a single attack, etc) but critical successes/failures with 'special effects' can serve that goal as well.

Full attacks don't have to take significantly longer than anything else to resolve as long as you're organized and use a calculator if you need it.

You're right about battles with foregone conclusions being boring as heck, though that's hardly something to lay on the humble attack roll.

It's also something that fumbles don't effect any more than enemy critical hits. Ultimately it's just a slightly worse result than the miss that was already gonna happen unless it's a -major- debuff like blindness, stun, or prone for multiple rounds (not that I'm aware of anything that causes that last one) except when it's already a tight battle.

In the case of an already tight battle the extra resolution will just drag things down and increase the PC's chances of ultimate failure, which is rarely a good thing.

Fumbles as an avenue for introducing more uncertainty -sounds- good in theory but the same thing can be accomplished by using enemies with special abilities.

There's also the matter that odds are pretty good that if a player rolled up something whose primary interaction with the enemy is mostly unmodified attack rolls with the occasional charge and/or power attack calculation, he doesn't -want- extra complications. Forcing that extra complication on -every- combat, regardless of how simple the foe is, seems a bit mean spirited, IMHO.

[Knaight]

It wasn't my uncle, it was one of his buddies. And it wasn't the first time it has happened to them either. I suspect alcohol is involved.

Alcohol usually is, and it doesn't mix well with firearms.

Coming back to gaming, if there was some sort of fumble system that only applied if the character was already severely wounded, or drunk, or drugged up on something other than alcohol - and it didn't have bizarre instant deaths where they make no sense - I'd like it a lot better. The idea that the expert swordsman swinging a sword around shortly after smoking a bunch of opium accidentally throws it away with some frequency works pretty well.

[Sartharina]

Sounding like a broken record here, but treating critical failures on attacks as Critical Defenses on Defense makes critical failures work.

But crit fails, ironically, are best when used on rarely-used skill checks, when the use of each is a dramatic moment.

[Kelb_Panthera]

Sounding like a broken record here, but treating critical failures on attacks as Critical Defenses on Defense makes critical failures work.

But crit fails, ironically, are best when used on rarely-used skill checks, when the use of each is a dramatic moment.

It only makes them -sound- better. Mechanically they're still just as bad.

As for skill checks, it's even worse; they're not normally subject to the natural 1 rule so you're adding a whole new vector for failure that never even existed before.

[Sartharina]

It only makes them -sound- better. Mechanically they're still just as bad.

No more than letting enemies attack at all.

As for skill checks, it's even worse; they're not normally subject to the natural 1 rule so you're adding a whole new vector for failure that never even existed before.

Which is why I said 'ironically' - they normally don't apply to skills, yet are actually pretty good on them (See: Shadowrun's 'glitches'. Or a substantial number of Digo Dragon's campaign quotes.)

[Prince Raven]

No more than letting enemies attack at all.

... what? I just don't understand. How is "the enemies actually fight back with all those swords and stuff they're carrying" just as jarring as "The mighty warrior, Ungoth the Lichbane, slayer of dragons and champion of the people drops his sword, trips over on a rock, or accidentally stabs himself roughly once or twice for every minute of combat".

[NichG]

Full attacks don't have to take significantly longer than anything else to resolve as long as you're organized and use a calculator if you need it.

I don't actually agree with this statement in general (its a bit too idealistic to assume that everyone is going to be this on-the-ball in a world filled with other people whose habits you can't personally control), but even assuming that its true and all actions have a constant-time resolution, the actual density of meaningfulness is still lower during attack sequences during e.g. the resolution of an attack that applies a status condition. Its basically because damage doesn't change the game state meaningfully unless it drops its target. The same would be true of rolling damage for an AoE spell (which is why I encourage taking average on damage dice pools, but that's another story)

It's also something that fumbles don't effect any more than enemy critical hits. Ultimately it's just a slightly worse result than the miss that was already gonna happen unless it's a -major- debuff like blindness, stun, or prone for multiple rounds (not that I'm aware of anything that causes that last one) except when it's already a tight battle.

In the case of an already tight battle the extra resolution will just drag things down and increase the PC's chances of ultimate failure, which is rarely a good thing.

Fumbles as an avenue for introducing more uncertainty -sounds- good in theory but the same thing can be accomplished by using enemies with special abilities.

There's also the matter that odds are pretty good that if a player rolled up something whose primary interaction with the enemy is mostly unmodified attack rolls with the occasional charge and/or power attack calculation, he doesn't -want- extra complications. Forcing that extra complication on -every- combat, regardless of how simple the foe is, seems a bit mean spirited, IMHO.

Well, if we take for example the system I proposed earlier in this thread, critical hits/misses are rolled once for the entire round and create a pool of points which any player can pull from during the round to create effects. In that system, a player who doesn't want the extra complication can just ignore the pool and let others spend it. Similarly, because its a pool which must be spent to have use, the critical successes/failures always have an element of decision-making to them so they encourage the players re-evaluating their plans to take advantage of a sudden opportunity (we have enough points this round to provoke a failed save!) or to protect themselves from a sudden disadvantage (the enemy has a lot of points this round, so better not let them get off a full attack on someone). So in that case its not about the immediate results of a roll being better or worse, but more about making it so each roll is an opportunity to consider using a shared luck-based resource. That said, the particular system I proposed would do a lot less for adding spice of this sort to a long attack sequence because I specifically made it to have a crit rate independent of how many attacks people get due to the 'higher level Fighter fumbles more often' and 'spellcasters often don't have to roll' problems.

[Sartharina]

... what? I just don't understand. How is "the enemies actually fight back with all those swords and stuff they're carrying" just as jarring as "The mighty warrior, Ungoth the Lichbane, slayer of dragons and champion of the people drops his sword, trips over on a rock, or accidentally stabs himself roughly once or twice for every minute of combat".

I'm talking "Critical failures are actually Critical Defenses" - As he's hewing through the goblin horde, Ungoth the Lichbane falters against Dies Horribly the Generic Goblin Warrior on the blood-soaked ground, who takes the opportunity to disarm Ungoth/Knock Ungoth Off Balance/force the terrible blade against its wielder in a miraculous (For the goblin) parry - buying the miniature fiends just a scant extra second before Ungoth recovers from the setback, and begins his slaughter with renewed vigor."

Or, "As Olaf One-Eye brought his axe down against the mighty dragon Numinex, the dragon caught the blade in his jagged scales, and his defensive twist manages to wrench it away from its wielder."

[Seerow]

I'm talking "Critical failures are actually Critical Defenses" - As he's hewing through the goblin horde, Ungoth the Lichbane falters against Dies Horribly the Generic Goblin Warrior on the blood-soaked ground, who takes the opportunity to disarm Ungoth/Knock Ungoth Off Balance/force the terrible blade against its wielder in a miraculous (For the goblin) parry - buying the miniature fiends just a scant extra second before Ungoth recovers from the setback, and begins his slaughter with renewed vigor."

Or, "As Olaf One-Eye brought his axe down against the mighty dragon Numinex, the dragon caught the blade in his jagged scales, and his defensive twist manages to wrench it away from its wielder."

This runs into the weird situation where enemies are going to more frequently get a stroke of good luck/a lucky hit in against characters with more attacks.

But I am mostly okay with the idea of if you roll a 1 on the first attack roll of a round, the enemy gets some action to inconvenience you (say a free disarm/trip/bullrush attempt). The question then becomes how do you apply a similar logic to the spellcasters? If the enemy rolls a natural 20 on their saving throw, does it cause some form of backlash on the caster? What about spells that are no save, no attack, just win? Is there some way for the caster to fail there? What about against archer type characters? If they fumble and there isn't someone next to them to take advantage, is there no side effect? Or do they get the special "I'm a clown" fumble chart while everyone else has their enemies causing problems for them?

Important things to consider.

[Sartharina]

This runs into the weird situation where enemies are going to more frequently get a stroke of good luck/a lucky hit in against characters with more attacks.

But I am mostly okay with the idea of if you roll a 1 on the first attack roll of a round, the enemy gets some action to inconvenience you (say a free disarm/trip/bullrush attempt). The question then becomes how do you apply a similar logic to the spellcasters? If the enemy rolls a natural 20 on their saving throw, does it cause some form of backlash on the caster? What about spells that are no save, no attack, just win? Is there some way for the caster to fail there? What about against archer type characters? If they fumble and there isn't someone next to them to take advantage, is there no side effect? Or do they get the special "I'm a clown" fumble chart while everyone else has their enemies causing problems for them?

Important things to consider.

For casters - critical saves can give either a backlash on the caster (Most single-target spells), or a boon (On AoEs). Archery is probably the 'safest', but also weakest option. Most no-save-just-suck spells tend to be weak for their level as well, IIRC.

[Kelb_Panthera]

I don't actually agree with this statement in general (its a bit too idealistic to assume that everyone is going to be this on-the-ball in a world filled with other people whose habits you can't personally control), but even assuming that its true and all actions have a constant-time resolution, the actual density of meaningfulness is still lower during attack sequences during e.g. the resolution of an attack that applies a status condition. Its basically because damage doesn't change the game state meaningfully unless it drops its target. The same would be true of rolling damage for an AoE spell (which is why I encourage taking average on damage dice pools, but that's another story)

A couple attack and damage rolls vs a saving throw and cyphering the change to the target's values vs opposed rolls and whatever resolution lies there (grapple, trip, loss of target awareness, etc); none of them take terribly long if the player and dm stay on the ball.

As for whether an attack can change the game state, that depends on how close things are. If things are tight, one missed attack can be the difference between having a pc survive the fight vs being dropped. Affecting how many rounds the enemy stays on its feet, and in a good fight there's always more than one enemy, is a achange to the game state, it's just a bit more subtle one than a debuff. Besides, unless you've got somesome kind of save or die, damage is necessary to resolve most combats, assuming you're not in the habit of leaving live foes behind you.

Well, if we take for example the system I proposed earlier in this thread, critical hits/misses are rolled once for the entire round and create a pool of points which any player can pull from during the round to create effects. In that system, a player who doesn't want the extra complication can just ignore the pool and let others spend it. Similarly, because its a pool which must be spent to have use, the critical successes/failures always have an element of decision-making to them so they encourage the players re-evaluating their plans to take advantage of a sudden opportunity (we have enough points this round to provoke a failed save!) or to protect themselves from a sudden disadvantage (the enemy has a lot of points this round, so better not let them get off a full attack on someone). So in that case its not about the immediate results of a roll being better or worse, but more about making it so each roll is an opportunity to consider using a shared luck-based resource. That said, the particular system I proposed would do a lot less for adding spice of this sort to a long attack sequence because I specifically made it to have a crit rate independent of how many attacks people get due to the 'higher level Fighter fumbles more often' and 'spellcasters often don't have to roll' problems.

While it's an interesting idea, it's not a fumble system. It's much more akin to some kind of mutant action point system than one for fumbles.

[Jay R]

Just one interesting fact to muddy up the argument:

This year's best-performing running back in the NFL has had more rushing attempts than any other, and has committed more fumbles than any other.

It's not totally out of line for a great fighter who makes more attacks to have more critical fumbles. In real life, it can happen.

[Prince Raven]

Seems like a bad comparision, he doesn't get swords stuck in him when he fumbles the ball, which explains why he's still alive despite all those fumbles. Under many critical fumble rules I've seen it is improbable for a character to reach even level 4 without accidentally chopping their own head off. And I'm sure even the worst fumbler in professional handegg fumbles somewhat less than a 5% chance every 6 seconds they have the ball.

[The Glyphstone]

Seems like a bad comparision, he doesn't get swords stuck in him when he fumbles the ball, which explains why he's still alive despite all those fumbles. Under many critical fumble rules I've seen it is improbable for a character to reach even level 4 without accidentally chopping their own head off. And I'm sure even the worst fumbler in professional handegg fumbles somewhat less than a 5% chance every 6 seconds they have the ball.

Though to play Asmodean Advocate here, professional fighters don't fumble 5% of every six seconds they carry their sword peaceably through a forest, only when they use it - the handegg analogy would be every time that player 'rolls' to catch a pass from the quarterback or to dodge a blocker/interceptor.




On-topic, my perspective on Fumbles has always been pretty much Seerow's rules posted earlier in the thread, though my primary pass/fail was the straw-dummy test. Fail that, and you can butter your fumble rules and eat them.

[Talakeal]

I'm talking "Critical failures are actually Critical Defenses" - As he's hewing through the goblin horde, Ungoth the Lichbane falters against Dies Horribly the Generic Goblin Warrior on the blood-soaked ground, who takes the opportunity to disarm Ungoth/Knock Ungoth Off Balance/force the terrible blade against its wielder in a miraculous (For the goblin) parry - buying the miniature fiends just a scant extra second before Ungoth recovers from the setback, and begins his slaughter with renewed vigor."

Or, "As Olaf One-Eye brought his axe down against the mighty dragon Numinex, the dragon caught the blade in his jagged scales, and his defensive twist manages to wrench it away from its wielder."

A good explanation for a warrior fumbling against a horde of weak enemies is that he hits them TOO hard and got his weapon stuck in the last one, meaning he needs a moment to pull it out. Of course, the DM may need to narrate this carefully as it is a side effect of past actions rather than a result of the action that is currently being fumbled.

Also, I hate iterative attacks in general. The math is annoying and slows down the game and they make combat really static with everyone having to just stand in place swinging. Fumbles are a good system made stupid by iterative (imo) and iterative is a bad system made worse by fumbles.

None of my homebrew systems use it, and when I run d20 I usually just give a damage bonus equal to BaB and ignore iterative attacks, allowing people with flurry of blows or duel wield to just attack twice on a standard attack.

[Stellar_Magic]

One thing to consider is to make critical failures 'optional'. If you're playing 3.5 with the 'hero points' rule from UA, you could give the player a choice of taking the fumble and gaining a hero point, or just missing.

I'm doing the same thing with a Star Wars Saga Edition game I'm running.

[oxybe]

@ Sartharina

If the problem is one of narration alone, where instead of "crit fail, you look like a fool" you get "crit fail, enemy looks competent" I wouldn't have much issue with it to begin with. As it currently stand in the normal failure rules, where a one is a miss and nothing else, this is perfectly acceptable and I would say preferred for most games.

The problem is when the GM starts inputting additional mechanical effects like "weapon is thrown across room" or "you hit your adjacent ally" where the game accidentally turns into one of black comedy, which is most often the result of a GM adding additional mechanics to crit failures "it's a failure so something bad happens to you".

I blame it on the name of the mechanic "critical failure": it lends itself well to the theme of spectacular failure by default. I'm pretty sure if it was called "automatic miss" instead of "critical failure" we wouldn't have as many charts where fighters hurl greatswords into ceilings or accidentally lob off the dwarf's foot or something.

But to reiterate, it's one thing to turn a 1 into a comedy of errors with narration but no additional mechanics involved, but people I've gamed with that gave any special attention to crit failures do so by adding a mechanical repercussion beyond "auto-miss".

It's a great idea I would like to see used more often when Crit Fails are pure narration, but i find that this is rarely the case.

[Sartharina]

There's not a problem with adding mechanics to critical failures.
In its most abstract, combat is on a 'victory' track.

A normal success advances the person down that victory track by one point.
A critical success advances the person down that victory track by two points, or advances the person down the track by a point and pushes the enemy back up the track one point.

A normal failure does not advance the person down the victory track.
A critical failure either sends the user back up a point on the victory track, or advances the opponent down the victory track by one point.

Critical failures are just as likely to occur as critical successes (Or even less likely), and are just as likely to be triggered by either side. While a player character will accrue more critical failures over the course of a campaign than a monster, he'll also accrue more critical success, and it's all irrelevant because only encounter-level statistics matter.

Losing a sword on your enemy's critical defense doesn't make you look any more incompetent than that same enemy disarming you on their attack. Other self-attacks are similar, but only really make sense on a defense (An enemy parrying so well as to drive the blade back against his assailant, or redirecting an attack to strike his assailant's ally instead). There tend to be awkward feats that allow this, but feats are just an obnoxious way of saying "No you can't do this".

[Jay R]

Seems like a bad comparision, he doesn't get swords stuck in him when he fumbles the ball, which explains why he's still alive despite all those fumbles. Under many critical fumble rules I've seen it is improbable for a character to reach even level 4 without accidentally chopping their own head off. And I'm sure even the worst fumbler in professional handegg fumbles somewhat less than a 5% chance every 6 seconds they have the ball.

Then the problem is the critical fumble table. It should almost always create an annoying problem, not a death.
Make a DEX check or drop your weapon.
You are open to the next attack. DEX defense does not apply.
You slip. Miss one attack.
Your shield strap breaks. No shield for the rest of the combat.

By the way, is that an absurd exaggeration, or have you actually seen entire parties destroyed by critical failures before 4th level?

[Prince Raven]

Entire parties, no. But I have seen one person go through 3 characters thanks to his ability to roll natural 1s and accidentally kill his own character. Only 1 non-caster in the original party made it to level 4.

You're right though, those are generally much tamer outcomes of a critical fumble (though not a huge fan of "drop your weapon"). My least favourite outcomes include: accidentally stab yourself, roll damage (yes, strength bonus, power attack, etc. did apply. Made me happy I was low strength and wielding a dagger, the 18 strength greatsword wielder was less impressed); accidentally hit adjacent ally (again, full damage); weapon breaks, you are now useless this entire combat; Lose all remaining attacks this round and fall prone/become flat-footed/drop your weapon/etc.

Yes, all of those happened in the same campaign. Critical fumbles only applied to mundanes and skill-monkeys, casters who didn't have to roll to hit and only ever used knowledge and social skills were completely unaffected.

[Jay R]

Entire parties, no. But I have seen one person go through 3 characters thanks to his ability to roll natural 1s and accidentally kill his own character. Only 1 non-caster in the original party made it to level 4.

I agree that in that case, the critical fumble table marred the game. But the problem isn't critical fumbles; it's a bad table.

You're right though, those are generally much tamer outcomes of a critical fumble (though not a huge fan of "drop your weapon").

It's not that bad an outcome, but I've never had a melee fighter without a backup weapon, even back in 1975 before I ever heard of critical fumbles. When skeletons show up, leave the sword in its sheath and draw the mace.

My least favourite outcomes include: accidentally stab yourself, roll damage (yes, strength bonus, power attack, etc. did apply. Made me happy I was low strength and wielding a dagger, the 18 strength greatsword wielder was less impressed);

On the extremely rare case that a "damage yourself" outcome occurs, power attack bonuses should not apply, because the sword isn't hitting at the spot all that power was focused on. A competent DM should also take into account the weapon. You can't stab yourself with a rapier, though you can cut your left hand with a really stupid parry. You can strike yourself with a sword, but not with the sweet spot; damage should be reduced.

accidentally hit adjacent ally (again, full damage);

This can happen, but should be ridiculously rare. It's happened to me once in 35 years of SCA melee. Again, it doesn't occur unless the ally moves in front or at the end of a spent swing. Lower damage.

weapon breaks, you are now useless this entire combat;

Draw your dagger, or punch, or kick, or tackle, or pick up rocks to throw. You are not useless.

Once you start reading about real warriors and what they take to battle, you'll see that there is always a backup weapon. Also, if you are fighting armed warriors, there are free weapons as soon as the first one dies.

Any player who thinks his character is useless if his primary weapon breaks has not yet begun to play the character of a melee fighter.

Lose all remaining attacks this round and fall prone/become flat-footed/drop your weapon/etc.

Becoming flat-footed and losing attacks is actually the most common "critical fumble" I've seen in SCA combat - often because something weird just happened that distracted you.

Falling happens. Roll, get up, or stab up.

Yes, all of those happened in the same campaign. Critical fumbles only applied to mundanes and skill-monkeys, casters who didn't have to roll to hit and only ever used knowledge and social skills were completely unaffected.

If they are out of the melee, they aren't getting constantly jostled. What makes it fair is that as soon as one single orc gets to melee range, the wizard is more helpless than the fighter who fell or dropped his first weapon.