Switching out the existing Playtest Critical System for Homebrew

[tivadar27]

I've given up on the existing critical system for the homebrew, because from all of the experiences I've had with it, the math behind it doesn't work. For the record, I'm going to the following system:

Rather than +/-10, it will now be based on your proficiency with the attack/ability.
You Critically Succeed... The Enemy Critically Fails...
Untrained No chance No chance
Trained: Natural 20 Natural 1
Expert: Natural 19-20 Natural 1-2
Master: Natural 18-20 Natural 1-3
Legendary: Natural 17-20 Natural 1-4
Note that you still must succeed/fail to critically succeed/fail.

This affects the following:
Skills: You critically succeed based on your skill training, critical failure is based on the task difficulty (easy->untrained, medium->trained, hard->expert, incredible->master, ultimate->legendary)
Saves: You critically succeed based on your saving throw proficiency, critical failure based on the proficiency of the ability/spell or level of the trap.
Attacks: You critically succeed based on your weapon proficiency, critically fail (where it applies) based on the target's armor proficiency.

The reasoning for this change follows:
* Due to balancing reasons, against challenges that are your level or higher, your chance of success is almost always 50% or lower. This essentially eliminates the critical system for challenging encounters, which is where it matters most.
* This allows both ends of the critical system to come into play a bit more often on any given roll, making for a wider array of outcomes.
* This is less math. If you roll really high, you likely hit, really low and you likely miss. The actual number on the die tells you if it's a critical or not.
* This gives more weight to actual proficiency outside of the +/-1, as they now dictate critical ranges.
* Easy tasks are now impossible to critically fail.

[Tholomyes]

I can see this idea. I've already posited that I think the 4 tier success system doesn't really work, as is, for skills and attacks, and I think that, especially for skills (if we are keeping the 4 tier success system), untrained characters not being able to critically succeed would do more for my sense of trained vs untrained, than the 1.3 fix of a flat -4. And this does, I think, solve some of the difficulty tuning, as tipping away from the +/- 10 system means you don't have to stick as closely to the 50% success metric, however there's something that I think the +/- 10 system does do that I think is valuable, which is that lower level threats are easier to critically succeed on (either in Skill DCs or attack rolls, or critical failures in opponents' saves). I don't know if that's valuable enough that it's better than the downsides of the +/- 10 system (and honestly, I'm leaning towards no on that one), but it does give me some pause.

[tivadar27]

Fair about the low-level threats. With regards to lower level skill challenges (let's say a level 1 hard when you're level 3), I'd actually say you should treat that as you would a level 3 easy task (2 levels lower), so it wouldn't improve critical success rates, but your chance of failing would be pretty small, and you'd have no chance of critical failure.

As for lower-level enemies, I'm not sure you need a higher critical rate to make things work. Lower health and a higher chance to hit really already does the trick here.

Honestly, I just don't know how to remedy the +/-10 critical system to make gameplay fun/interesting and allow for a higher chance of success for someone who's "good" but not "great" at a task. Hence moving away from it.

[iNickedYerKnickers]

I don't understand why it would get easier to crit fail with better proficiency. That's counter-intuitive. The more proficient one is at a thing, the less likely he/she is to fail at it.

Otherwise, I like the concept and will suggest it to my group.

[Darksol the Painbringer]

I think the comparison is "Player Dice/Enemy Dice."

So, if a player makes an attack with Legendary Proficiency, for example, they critically succeed automatically on a 17-20, whereas if a bad guy rolls a 1-4 on a saving throw from a spell or effect with Legendary Proficiency, it's an automatic Critical Failure.

If anything, for things like skill checks, the second row should be invereted; you can never critically succeed on untrained skills, but you will critically fail on 1-4. Whereas Legendary proficiency in skills means you critically succeed on a 17-20, but can't ever critically fail.

[PCScipio]

I agree that the+/-10 critical system seems to be having a lot of unintended consequences.

[Lady Melo]

PCScipio wrote:

I agree that the+/-10 critical system seems to be having a lot of unintended consequences.

I feel like it is definitely showing it's side effects quite a bit, however I think a lot of that could calmed down by the system simply dialing down some of the effects of Criticals and excepting that they are okay to have. In some cases they seem really forced in there as well.

[tivadar27]

Darksol the Painbringer wrote:

I think the comparison is "Player Dice/Enemy Dice."

So, if a player makes an attack with Legendary Proficiency, for example, they critically succeed automatically on a 17-20, whereas if a bad guy rolls a 1-4 on a saving throw from a spell or effect with Legendary Proficiency, it's an automatic Critical Failure.

This is basically correct. Sorry if the summarization table was a bit unclear, but hopefully the concept is understandable.

Darksol the Painbringer wrote:

If anything, for things like skill checks, the second row should be invereted; you can never critically succeed on untrained skills, but you will critically fail on 1-4. Whereas Legendary proficiency in skills means you critically succeed on a 17-20, but can't ever critically fail.

As for skills, I decided to do successes based on the die roll (so you're right, with Legendary, the intent was you critically succeed on an 17-20), but critical failures are based off the difficulty of the task (assuming it's of your level), so "High" is a failure on 1-2.

I did this so that tasks that are "Trivial" (no failure chance) you don't even have to roll on if you already can't fail, and can assume success. The old system had a big problem where, for example, if you had a +15 and were trying a trivial level 1 task (DC 10), you'd still fail on a natural 1. With no "auto" dice rolls anymore, if a task is so trivial that there's no chance of failure and you're already good at it, there's also no chance of critical failure.

[The Once and Future Kai]

I really like the +10/-10 critical system but it does seem responsible for the tight math. Maybe if criticals were weakened it would be viable?

[tivadar27]

The Once and Future Kai wrote:

I really like the +10/-10 critical system but it does seem responsible for the tight math. Maybe if criticals were weakened it would be viable?

Yeah but the problem with that is that people are *already* upset about infrequent/weakness of spells...

I honestly really thought the +/-10 system was a cool idea coming in, but I've seen how it affects the game and just came to the conclusion that it's fatally flawed. At least now I can have an enemy that a fighter hits 75% of the time on their first hit (other classes around 60) and not have it trivially easy due to criticals.

[Mad Master]

Actually, both the "new ideas" about how to assign critical hits/misses are heavily flawed. Besides, they are not new at all, since they were both used in the revised 2nd edition of Advanced Dungeons & Dragons.

Having an automatic critical hit on 20 and critical miss on 1 brings up a really awkward situation: an highly skilled character would make less critical hits per successful attacks than a poorly skilled one, with the extreme chance that the poorly trained one could even reach a 100% critical to hit ratio. That would mean that every successful attack of the poorly trained guy would be a critical hit, while could happen that only 1 out of 19 successful attacks on average would be a critical hit for the highly trained guy.
The reverse is true for critical misses, which again is bad, since it could happen that when the highly trained guy fails, he always critically fails, while the poorly trained guy would critically fail only sparingly.

The margin of success has the correct maths: the highly trained should score more critical hits and less critical misses than the poorly trained guy. But the real issue with this system is that if the foes are balanced, there could be no chance for critical hits or misses, especially with a margin of success as large as 10.
Basically, if you hit with an 11 or more, you have no chance for a critical hit, while if you hit with a 10 or less, you have no chance for a critical miss.
Having a system that tries to balance all that it can doesn't help too.

Those issues were solved in the passage from AD&D 2nd Edition to D&D 3.0. The extremely simple solution was the confirmation roll, whose only fault was in the naming: it should have been called something like "critical roll", as a followup of a particularly successful "to-hit roll", to create less expectations and frustrations.