Confirming crits, good idea, bad idea, or worst idea ever

[the Jester]

I like (and miss) crit threat ranges. I dislike confirming crits. To hell with realism, I want my 20 to mean something!

This.

 

[Kingreaper]

Confirming crits=annoying.

Now, if nat 20=crit, and you then roll to see if it's a minor or a major critical, could be an interesting mechanic, but nat 20 needs to be "Cool I crit!" not "Cool I... Oh, wait, confirmation roll... oh, a 5, well I guess I just hit then :-("

 

[mkill]

The weird part wasn't just crit confirmation, it was also the weird dance around getting rid of it with a feat or class feature. It was just not possible, all you could get was a +4 on the roll IIRC. 3E just had too many fiddly bits in all places. It's like one of those swiss army knife with 30 tools - sure, it promises that you can do anything, but try to do any real work with it and it breaks in your hands.

My top 3 of "3E fiddly bits I houseruled out, never missed and never want to see again"

* cross-class skills cost 2 per rank
* favored class
* crit confirmation

Funny that PF kicked #1, left #2 in but made it bearable and kept #3

 

[Crazy Jerome]

It should also be noted that if you are absolutely wedded to the concept of directly simulating "people with more accurate attacks critical more often, however swingy those criticals are," then confirmation rolls is probably your best mechanic.

However, if what you want is the concept of "people with more accurate attacks do more damage on criticals," then you can get the same effect with less annoyance and handling time by simply adding an extra damage roll for crits, based on the flat plus of the attack. This is especially true in a system that seeks to flatten out the attack bonus.

That is, you don't need to consider the relative accuracy of the attack versus the AC (or other defense). If Joe the 1st level Fighter has +4 to hit with his longsword, he gets some modest bonus to damage on all crits (in addition to max damage or normal roll or whatever else you do to make them special). Joe at 10th level hits harder, so that number goes up.

Except for one special case, it gets tricky to include special modifiers (like +2 to flanking) in that equation, but some might see that as feature, not bug. You get bonus damage on criticals based on your raw accuracy--separate from the situation. Of course, if you define the system where the bonus to damage on crits equals the modifier you had with the attack, then it isn't that hard to include the situations, either. Flanking becomes +2 to hit and +2 damage on crits, every time. (Such a system would presumably keep the base damage relatively low.)

 

[Estlor]

I'm not a fan of rolling to confirm crits. I prefer the simplicity of 4e where 20 = auto-hit, if it hits anyway it's a crit.

That said, they're a little more palatable under the model of the confirmation roll only being for the BONUS damage. The fact that a nat. 20 still gets you an auto-hit and max damage makes it meaningful even if you fail to confirm.

 

[delericho]

I would really prefer that natural 20 mean -- "max damage on current roll" + "extra damage roll based on whatever weapon, magic, and special crit options you have with that attack" --then with options to replace the second part of that with a variety of effects (if you want). For one thing, it's a lot easier to manipulate the options to get the style that a table wants.

Too powerful for my taste. Way too powerful for my taste.

If we're having 20 auto-crit, then a critical should simply be "max damage (inc bonuses)". Even the 4e effects that do even more damage on a crit should be eliminated.

Only if we're having a confirmation roll should we have "roll double damage" or "max damage plus even more damage" or "roll damage plus effect".

Having a 20 auto-crit already makes crits a bit too common. I would agree that that's fine because of the emotional argument that "a 20 should be special"... but if we're going that way then we should make sure we don't make a 20 too special.

IMO, of course.

 

[Bohemian Ear Spoon]

it seems like rolling nat 20s still only happes a couple times a session, how is that too common? My theory is that the fuzzy math involved in confirming crits has no bearing in real world game play.

on the other hand, critting on anything other than a nat 20 blows my argument out of the water.

On the third hand, I would be content with a crit being auto max damage and then a roll to confirm "extra bad-assery"

 

[Crazy Jerome]

Too powerful for my taste. Way too powerful for my taste.

If we're having 20 auto-crit, then a critical should simply be "max damage (inc bonuses)". Even the 4e effects that do even more damage on a crit should be eliminated...

Part of my assumption that this structure is able to cater to a wider variety of styles is that it allows you to turn that "extra thing on crit" into "nothing" easily, if you so choose. Max damage on a natural 20 is a relatively modest base from which to do that. Or more likely, you use the normal rolled damage + some relatively modest extra on critical.

You can't so easily manipulate something like the 3E great axe crit multiplier, because it has such a wide range of effect for a relatively narrow set of parameters.

 

[Scribble]

I miss the old days where you had a 5% chance of getting your head lopped off.

 

[delericho]

it seems like rolling nat 20s still only happes a couple times a session, how is that too common?

Well, auto-crit on a nat-20 is a flat 5% of all attacks resulting in a crit (barring the corner case of the very-high AC... but frankly, I'd be inclined to just drop that rule). With five PCs and five monsters, that's an average of one crit every other round or so.

Basically, yeah, I feel that that's "too common". I would be much happier with the 2-3% chance that the confirmation roll gives.

Of course, it's all a matter of taste. YMMV, and all that.