Critical Hits

[Ahnehnois]

1d8+1 isn't all that rare (for example, a 12 strength Weapon Finesse rogue in 3e, or a fighter with a strength of 16 or 17 in 1e/2e). Besides, the same argument could be made with 1d8+4 or 1d12+8. Both of those are also lower on a min roll, after doubling, than the max. Rolling minimum on a 1d8 or 1d12 isn't exactly a rare occurrence.

Also, there are a fair number of attacks that don't get a modifier at all. The 3.5 version of disintegrate is a prime example. Rolling low damage on a crit with that spell is lame (at least compared to what you could inflict with an average non-crit).

Like I said, I don't really care if the low-damage characters have weak crits occasionally. I've never had a player complain about that. If you roll two large dice and roll low, it feels like you deserve low damage. That said, a good fighter or barbarian will have a damage modifier equal to or greater than than the dice of his weapon by the time the wizard starts fireballing, and they are the ones who roll the most attacks (and therefore who the crit system is really for).

Disintegrate isn't really the main thing crits should be designed around, but realistically 3.X disintegrate rolls so many dice that the odds of a crit being worse than a normal hit are infinitessimal.

In a broader sense, damage in D&D is always random, non-d20, and dissociated from attack. Even independent of crits, you can roll a 19 on your attack and a 1 on damage. That is a bit of a downer, but the occasional bad roll is part of the drama of the game.

That's why the 4e way is that you get bonus dice on top of the max. High crit and magical damage bonuses all add dice. The only thing I think 4e could have done better in this respect would have been to grant an inherent crit bonus at 1st level.

So basically, what you're saying is that maxing the damage is not (by itself) sufficient. You're also saying something that wasn't in the original poll options. So basically, we agree on this.

I don't think ranges/multipliers are a good way. Multipliers cause damage to get way out of hand. x2 isn't bad, but x3 and x4 make the game too swingy and luck based. Either you have to make PCs too tough, able to withstand at least 5 regular hits, or you end up with characters who get taken out by a lucky (cheap) shot before they can even act.

Ranges enable the above strategy and also cheapen crits (because, IMO, they make them too common). I'm not saying that we should never see a range of 19-20, but I don't think it should be common, and we definitely shouldn't see ranges of 12-20 ever again.

Multipliers are swingy. Swingy is not necessarily a bad thing. It makes combat exciting and scary and pushes balance towards people who are using weapons and using them well (casters are almost always 20, x2). Being killed by a "cheap" crit is really no worse than an SoD (which is of course it's own issue).

I'm guessing you wouldn't like my rogue revision that makes sneak attack into an improved threat range. My rogue maxes out at critting on a 2 when they've lost dex to AC (with a keen rapier and improved crit) and having +2 to crit multiplier, and a little bonus damage. It makes rogues really scary (and takes away the halfling stabbing you in the eye for 10d6 damage factor).

I think altered crit parameters should be common among skilled characters, and are one significant way of balancing the non-casters.

4e uses the properties High Crit (roll the weapon's damage die and add it to any crit) and Brutal (reroll the weapon's damage die if it is equal or below the brutal number). Those properties differentiate weapons too, but without making them one-shot kills.

Those aren't bad, but I don't think they replace the power or flexibility of crit-based properties. I'd be happy to see additional variables like these add to the diversity of weapons or combat styles.

That's good hear. At least we can both agree on that! Perhaps there's hope for DDN after all!

This may be an area where a combined system can satisfy a lot of people. Here's to hope.

 

[Fanaelialae]

Like I said, I don't really care if the low-damage characters have weak crits occasionally. I've never had a player complain about that. If you roll two large dice and roll low, it feels like you deserve low damage. That said, a good fighter or barbarian will have a damage modifier equal to or greater than than the dice of his weapon by the time the wizard starts fireballing, and they are the ones who roll the most attacks (and therefore who the crit system is really for).

Disintegrate isn't really the main thing crits should be designed around, but realistically 3.X disintegrate rolls so many dice that the odds of a crit being worse than a normal hit are infinitessimal.

In a broader sense, damage in D&D is always random, non-d20, and dissociated from attack. Even independent of crits, you can roll a 19 on your attack and a 1 on damage. That is a bit of a downer, but the occasional bad roll is part of the drama of the game.

So basically, what you're saying is that maxing the damage is not (by itself) sufficient. You're also saying something that wasn't in the original poll options. So basically, we agree on this.

Multipliers are swingy. Swingy is not necessarily a bad thing. It makes combat exciting and scary and pushes balance towards people who are using weapons and using them well (casters are almost always 20, x2). Being killed by a "cheap" crit is really no worse than an SoD (which is of course it's own issue).

I'm guessing you wouldn't like my rogue revision that makes sneak attack into an improved threat range. My rogue maxes out at critting on a 2 when they've lost dex to AC (with a keen rapier and improved crit) and having +2 to crit multiplier, and a little bonus damage. It makes rogues really scary (and takes away the halfling stabbing you in the eye for 10d6 damage factor).

I think altered crit parameters should be common among skilled characters, and are one significant way of balancing the non-casters.

Those aren't bad, but I don't think they replace the power or flexibility of crit-based properties. I'd be happy to see additional variables like these add to the diversity of weapons or combat styles.



This may be an area where a combined system can satisfy a lot of people. Here's to hope.

I'd say (IMO) that maximized crits are necessary but not sufficient in themselves. Of course, the poll names 4e for the maximized approach, which uses maximum plus extra for almost the entire span of the game (as soon as you acquire magical weapons, or immediately if you have High Crit).

As to your objection that this only occurs with weak combatants, not so. 1d12+8 isn't weak. Yet a minimum roll (9), doubled is 18. The maximum non-crit of 1d12+8, however, is 20. I don't have a problem with rolling a 1 for damage on a (non-crit) nat 19, but I think a crit should be special. It should never be a let down. Maximum damage guarantees that, because it's still pretty exciting to roll max damage for a regular hit. Getting that on a crit should be the least a crit can do.

Multipliers add too much swing to the game. The comparison to SoD is misleading. SoD can be easily made modular through several approaches. Making a crit multiplier the default makes it much harder to remove from the system cleanly. Not everyone wants SoD in their game, so having a crit system that is effectively an equivalent would be a bad design move.

Also, the 4e approach allows a much smoother, graduated increase than multipliers allow for. A system with multipliers jumps. You have x2, and the smallest incremental increase after that is x3, then x4, and so on. The only other means to increase the crit is to increase normal damage dealt (not always desirable). The max+ approach allows you to increment it at any rate you choose. You still get a bigger crit if you increment normal damage, but if you want to increment just crit damage you can add +1 (or +1d6, or +1000d10) damage just to crits. Especially if the crit adds dice, you can still get the excitement of rolling a high crit (rolling high on your crit dice) but without the disappointment of a crit that's not really a crit (because you're always guaranteed to do more than a regular hit).

Yeah, you're right, I would not like a rogue who crits on a 2+. I think making them too common cheapens crits. It takes the excitement out of them when crits are almost as common or more common than regular hits. I'm of the opinion that crits should be special.

Indeed, here's to hope!

 

[Ahnehnois]

As to your objection that this only occurs with weak combatants, not so. 1d12+8 isn't weak. Yet a minimum roll (9), doubled is 18. The maximum non-crit of 1d12+8, however, is 20. I don't have a problem with rolling a 1 for damage on a (non-crit) nat 19, but I think a crit should be special. It should never be a let down. Maximum damage guarantees that, because it's still pretty exciting to roll max damage for a regular hit. Getting that on a crit should be the least a crit can do.

Of course, the odds of you rolling d12 and rolling two 1's are 1 in 144, and the odds of getting a 1 and a 2 aren't much higher. It's a very rare event that a crit would be that weak, one that I don't think is a serious design consideration. Conversely, what are the odds that you get better than a max damage crit, or even better than max damage plus (let's say) 1d12 on top? Much higher. For a damage-multiplied crit, the ceiling is almost always much higher, which I think is more than worth having the floor be a little lower. (Of course in the proposed combo system, you get both a high ceiling and floor on a crit).

Multipliers add too much swing to the game. The comparison to SoD is misleading. SoD can be easily made modular through several approaches. Making a crit multiplier the default makes it much harder to remove from the system cleanly. Not everyone wants SoD in their game, so having a crit system that is effectively an equivalent would be a bad design move.

I am of the opinion that instantaneous death should always be on the table. There are opt-outs for both SoD and powerful crits that aren't terribly difficult to implement if you want to take it off.

Not having multipliers, conversely, kind of hamstrings the design. Without them, fighters, rogues, and their nonmagical brethren have to be looked at differently, especially the rogues and other finesse combat characters.

Also, the 4e approach allows a much smoother, graduated increase than multipliers allow for. A system with multipliers jumps. You have x2, and the smallest incremental increase after that is x3, then x4, and so on. The only other means to increase the crit is to increase normal damage dealt (not always desirable). The max+ approach allows you to increment it at any rate you choose. You still get a bigger crit if you increment normal damage, but if you want to increment just crit damage you can add +1 (or +1d6, or +1000d10) damage just to crits. Especially if the crit adds dice, you can still get the excitement of rolling a high crit (rolling high on your crit dice) but without the disappointment of a crit that's not really a crit (because you're always guaranteed to do more than a regular hit).

Depending on the exact numbers, you still risk losing the specialness of crits. A mid-level fighter who crits on a full PA with his axe for X3 damage could easily be looking at rolling, say, 1d8+20 three times. A couple of extra dice is pretty trivial compared to that. That's not to say that bonus damage on crits doesn't have a role, but the multiplier is precisely where the 'wow' factor of a crit comes from.

Those 2e backstabbing rogues were really great. A 3e barbarian two-handing a greataxe is pretty spectacular.

Indeed, here's to hope!

 

[Fanaelialae]

Of course, the odds of you rolling d12 and rolling two 1's are 1 in 144, and the odds of getting a 1 and a 2 aren't much higher. It's a very rare event that a crit would be that weak, one that I don't think is a serious design consideration. Conversely, what are the odds that you get better than a max damage crit, or even better than max damage plus (let's say) 1d12 on top? Much higher. For a damage-multiplied crit, the ceiling is almost always much higher, which I think is more than worth having the floor be a little lower. (Of course in the proposed combo system, you get both a high ceiling and floor on a crit).

The game before our most recent game, I rolled three nat 1s in a row. The odds of that are 1 in 8,000, or 0.0125%. Since that's looking at groups of 3d20, that should only occur once for every 24,000 rolls of a 1d20. During our most recent game, the player who sits across from me rolled three nat 1s in a row. I can tell you with absolute certainty that we did not make 8,000, much less 24,000 rolls between the former and the latter session. Yet it happened nonetheless. And it's hardly the first time we've seen something like that.

I think if it can happen, the designer should at least try to take the eventuality into consideration. I, and everyone I know (who played 3e), have felt the disappointment of "crits that weren't crits". Heck, I've personally witnessed multiple Disintegrate "crits that weren't crits" (we had a sorcerer who loved that spell).

Strings of bad rolls happen. You can't just look at the mathematical likelihood of something, because the way it comes out it play will only reflect that if you look at a huge pool of data. On the micro scale, such as the individual game session, you can and will have significant deviation from that norm.

I am of the opinion that instantaneous death should always be on the table. There are opt-outs for both SoD and powerful crits that aren't terribly difficult to implement if you want to take it off.

Not having multipliers, conversely, kind of hamstrings the design. Without them, fighters, rogues, and their nonmagical brethren have to be looked at differently, especially the rogues and other finesse combat characters.

Depending on the exact numbers, you still risk losing the specialness of crits. A mid-level fighter who crits on a full PA with his axe for X3 damage could easily be looking at rolling, say, 1d8+20 three times. A couple of extra dice is pretty trivial compared to that. That's not to say that bonus damage on crits doesn't have a role, but the multiplier is precisely where the 'wow' factor of a crit comes from.

Those 2e backstabbing rogues were really great. A 3e barbarian two-handing a greataxe is pretty spectacular.

Not having multipliers doesn't hamstring the design, it just takes it in a different direction. It's still very easy to make SoD crits. Just give everyone +5d20 crit damage, for example.

How do you, on the other hand, implement a solution for people who want crits but not SoD crits, if you have multipliers?

Give them bonus hp? Well now you have the problem that you've made them tougher against non-crits, which isn't necessarily what you were looking to accomplish.

Give them DR against crits? Well now you've nerfed the low end of crits more than you may have wanted to.

A multiplier has been the traditional way, but it isn't the best way (IMO). As I stated before, multipliers jump. If you want to have a dial for crit lethality, max + additional damage is the way to go. Someone who desires low lethality for crits can rule that bonus dice always deal minimum damage. Someone who desires high lethality, can rule that bonus dice are always maximized, or if he prefers something more random simply add any number of additional dice to the mix.

The problem with a multiplier is that it's based upon damage. You can't really change crits in that type of system without changing damage, or introducing something convoluted (like a crit damage penalty).

The advantage of critical dice is that they're independent. Modular, if you will. Damage does factor in (because it's maximized) but you can add any amount of crit damage (or reduce crit damage) on top of it without influencing non-crit damage. It gives you much more control. Instead of a dial that just reads Low/Medium/High, you have a dial that starts at zero and goes to infinity.