I have long been dissatisfied with the basic KAP dice mechanics. The problem is the way crits and fumbles are handled. To briefly recap here:

A. A Critical Success is scored when you roll your modified skill level exactly.

So far so good. I mean, rolling the exact number is a little weird in that it presents a moving target, but it doesn't really present any real cognitive problems.


B. A Fumble is normally scored if you roll a natural 20 — unless your modified Skill level is 20 or more — in which case see C, below.

Foul! This is not only counter-intuitive to the way many gamers think (a natural 20 is usually a crit) it's counter to most logic (a 1-19 is good, a 21 is good, but a 20 is terrible) and it introduces a cumbersome exception. In fact, it introduces an exception, ​a conditional, and a reversal to the normal rule in one fell swoop. CLUNK!

C. If your modified skill is more than 20, for every point over 20 in the Skill add this number to the die roll (an exception to the “never add to your roll” rule). A 20 or more results in a Critical, as always. Also, characters with 20 Skill level can never Fumble (another exception to remember).

Fie! This introduces another conditional, an additional computation, an exception to a standing rule, and a mechanical flaw (everyone should have occasion to fumble, IMO). DOUBLE CLUNK.

An additional challenge is the fact that rules as written as just about impossible to emulate in a virtual tabletop environment, like Roll20, where you simply can't emulate KAP rolls. Surely, there's a more elegant, modern way to handle crits and fumbles.


I'll start with a possible solution:

For unopposed rolls, why not just roll d20 and add your modified skill with a Target Number of 20 (originally suggested by Thijs Krijger) — if you roll 20 or more than your modified skill (basically, a natural 20) you crit. If your modified skill + d20 is 0 or less, you fumble. Done.

For opposed resolution, compare your d20 + modified skill to the opponents’ roll + modified skill. If your roll is 20 more than his, you crit. If your roll is less than zero, you fumble (you couldn't compare vs. the opponents modified roll, otherwise every time they crit, you also fumble, which seems too harsh).

This may not be the best solution — and I'm not even sure that Roll20 could emulate it (I've not tried yet) — but I offer it as a conversation starter.


Best,


T.

Frankly, I have no problem with any of the points you raised. :-) Perhaps because I've played with stranger mechanics so rolling a d20 fumbling on a natural 20 does not raise any eyebrows. My only problem with the rules is the way the game bogs down when an opposed roll is guaranteed a critical result on both sides. The system is not clear about this: if both oponents roll a critical, who wins? Sword breaks ties with other weapons, but what happens when it's a Spear against a Mace or an opposed Horsemanship roll? I tend to consider both results as normal results and pick up from there.

The system is not clear about this: if both oponents roll a critical, who wins?

Both count as a 20, so it is a draw. Other weapons than swords do not break other weapons on a draw, so nothing happens. Greg added a rule that a critical draw causes 1d3 damage (armor doesn't protect), to reflect the slow chipping down of two highly skilled knights fighting each other.

The way we do it in our house rules is that we simply let criticals cancel out, like I think you do, too. So say a skill 25 rolls a 19, that becomes 19+5 = 24, a critical. Skill 15 opponent rolls 15, which is a critical, so it counts as 20. Both are criticals, so that aspect is cancelled, and we compare only the numbers: 24 vs. 20. 24 is larger, so it is a normal success, and 20 a partial success. I guess you could take it one step further and reduce the 15 roll back to 15 (since you cancel the critical), but we haven't done that.


As for the dice mechanics, I LIKE the KAP compare-rolls mechanic. I find it extremely fast and easy, even in comparison to 1d20+skill+modifier (especially since that is often two teens summed together, which is much more of a pain than skill+5, which you have to do anyway). The only tweak that I think might make things even easier would be to change critical to a roll of 20, regardless of skill, and make 1 a fumble. This means you wouldn't have to change the rules for skill 20+ when it comes to fumbles and criticals. But in my personal experience, the players have picked up pretty quickly the idea of 'roll under your skill is a success, and exactly your skill is a critical', too. Then again, none of us come from D&D 3rd edition, either. I can see why those people would have it easier if the rolling mechanism stays the same. One more thing to add is that Pendragon is essentially d% system from Runequest, just mapped to 5% increments (d20).

EDIT:
Just to add...

In a 1d20+skill, you add an additional element (2a). This is how the comparison goes, in my mind:
1) Roll 1d20. This is done in both.
2a) In 1d20+skill, calculate the addition, which often is a high number added to another high number. In KAP, skip this. Overwhelming advantage, KAP.
2b) In 1d20+skill, compare the result with 20. In KAP, compare result with skill. Slight advantage, 1d20+skill, since it is not a moving target number.
3) If opposed, compare the successful results to one another. In 1d20+skill, these numbers are invariably over 20, or they wouldn't be successes in the first place. In KAP, you are comparing two relatively low numbers, which is faster. Advantage, KAP.

In short, 2b and 3 are definitely at worst a wash, leaving 2a) as the overwhelming disadvantage for the 1d20+skill system.

The gripes about the criticals and fumbles and flipping... Well, those happen in a minority of cases, where as 2a would be needed in all the rolls, including all the NPC rolls. GM_Morien does not want to calculate that. And if that is felt to be a big enough problem, I would fix it simply by making 20 a critical and 1 a fumble, as I said in above. This would also take care of the 'criticals count as 20' rule, since now they naturally would be counted as 20, too. Also, I note that Taliesin's solution for the criticals adds ANOTHER comparison (i.e. more complexity), to see if your roll is 20 or more than the opponent's roll.

As for Roll20 not being able to do KAP system, sounds to me that is something to bring up with Roll20 and get them to add another dice resolution mechanic, rather than foisting d20+skill on the rest of us. Or simply roll d20 like the rest of us peons and calculate the result yourself. :)

Let me lead off by saying that I very much like the Pendragon mechanic, and consider it extremely elegant.

That said, on your system, I think that you are actually exacerbating some of the problems that you mention. For example, it's much harder to fumble by your rules because you need to roll zero or less. That means that you need a penalty to the roll at least 1 greater than your skill. So, by normal rules, a -1 penalty would allow a Skill 20 knight to fumble, but by your rules, he would need to be getting a -21 penalty.

If you want to as directly as possible emulate KAP while switching to an always roll high system, try:

Roll 1d20+Skill. You succeed if the total is at least 21. You crit on a natural 20 (if you also succeed) or if the total is at least 40. You fumble on a natural 1 (if you also fail). For an opposed roll, the higher roll always wins (if it succeeds), except that a crit beats an ordinary success, and two crits tie.

Now, you could modify from here and say that a 1 always fumbles or a 20 always crits if you liked, but this should model basic KAP probabilities much more closely (I think exactly, but I'm not certain).

The KAP game system is in concept very different than most other games. While most other games roll a die against a fixed number, KAP basically roll opposed.
Games, like D&D, use the principle higher is better, but that is in a KAP game not the case. So there is no reason to assume that 20 is better than 1 as the result of your roll is always compared to another roll, and as such even a 1 can beat a 20.

The problem with the KAP system occurs when values are higher than 20. Then a calculation is needed and when skills rerach 30 or higher the system goes totally wonky. Also when both competitors have a skill higher than 20 the system creates a problem.
In my game I have made skills hiogher than 20 more difficult to get. Increasing skills above 20 you must either roll a succesful check in the winterphase. Glory points can only used, but you need new value-20 in glory points to increase it by 1. So to get a score of 22, you need 2 glory points to reach.
Also the option Morien uses (crits cancel each other out) is also a nice way to work with. Haven't used it myself , but are considering it.

By the way I use a different method in calculating the critical range. Instead of adding the (skill-20) to the roll we calculate 20-(skill-20). All rolls higher than that value are criticals. So a skill of 25 would mean all rolls of 15 and higher mean a critical (20-(25-20)=15)

I'd much prefer:
D20 + skill + modifiers: 20+ = success, 40+ is crit, <1 is fumble..
Great thing, is that you CAN also have 30+ as 'partial crit'.. and with opposed high skill rolls you CAN crit if 10+ or 20+ higher than opponent instead
So much more elegant and simpler..
I play with groups of PhD rocket scientists and they have problems grasping old system quickly..

The thing I dislike most old system is that a single +1 or -1 is the difference between crit and fail.. [or even fumble]
Player: I rolled a 20 and I have skill 20, that's a crit!
GM: No, remember that -1 because of your scratched sword... you fumbled...

As for the dice mechanics, I LIKE the KAP compare-rolls mechanic. I find it extremely fast and easy, even in comparison to 1d20+skill+modifier (especially since that is often two teens summed together, which is much more of a pain than skill+5, which you have to do anyway).

That's a good point. I hate math already! But in a digital world, you need not do math at all. I understand tht not everyone cares about playing in a digital world, however. But it's a shame when the mechanics are so convoluted that they can't be accomplished with scripting.


The only tweak that I think might make things even easier would be to change critical to a roll of 20, regardless of skill, and make 1 a fumble.

But then someone with a modified skill of 30 Sword, would have then same chance to fumble or crit against an opponent with an adjusted score of -5. Doesn't seem right. You're basically saying everyone would have a 10% chance of critting or fumbling in any round. Also doesn't seem right to me. Makes skill count for much less, and randomization count for much more. One could argue that anything can (and does) happen in a fight, but it still seems to penalize extraordinary characters and reward wimpy characters.


This means you wouldn't have to change the rules for skill 20+ when it comes to fumbles and criticals. But in my personal experience, the players have picked up pretty quickly the idea of 'roll under your skill is a success, and exactly your skill is a critical', too. Then again, none of us come from D&D 3rd edition, either. I can see why those people would have it easier if the rolling mechanism stays the same. One more thing to add is that Pendragon is essentially d% system from Runequest, just mapped to 5% increments (d20).

I don't come from D&D, either, and did not mean to draw comparisons between that system and KAP in any way. I'm just trying to figure out a way to reduce what I regard as some creaky dice mechanics in KAP.


EDIT:
Just to add...

In a 1d20+skill, you add an additional element (2a). This is how the comparison goes, in my mind:
1) Roll 1d20. This is done in both.
2a) In 1d20+skill, calculate the addition, which often is a high number added to another high number. In KAP, skip this. Overwhelming advantage, KAP.
2b) In 1d20+skill, compare the result with 20. In KAP, compare result with skill. Slight advantage, 1d20+skill, since it is not a moving target number.
3) If opposed, compare the successful results to one another. In 1d20+skill, these numbers are invariably over 20, or they wouldn't be successes in the first place. In KAP, you are comparing two relatively low numbers, which is faster. Advantage, KAP.

In short, 2b and 3 are definitely at worst a wash, leaving 2a) as the overwhelming disadvantage for the 1d20+skill system.

But waitasecond. This analysis completely ignore my main issues with the KAP system, which is the additional cognizant load of

1.) Remembering that your crit number can change with every round, depending on modifiers
2.) Remembering several "exceptions to the rule" — those are recounted in my original post — and a couple of reversals in the internal logic of the system.

Is the cognitive load imposed by the one additional calculation more or less than the cognitive load of remembering and applying the pretzel logic described above? That is the key question (to me).


The gripes about the criticals and fumbles and flipping... Well, those happen in a minority of cases, whereas 2a would be needed in all the rolls, including all the NPC rolls. GM_Morien does not want to calculate that. And if that is felt to be a big enough problem, I would fix it simply by making 20 a critical and 1 a fumble, as I said in above. This would also take care of the 'criticals count as 20' rule, since now they naturally would be counted as 20, too. Also, I note that Taliesin's solution for the criticals adds ANOTHER comparison (i.e. more complexity), to see if your roll is 20 or more than the opponent's roll.

Well, you're already comparing roll in opposed resolution anyway, and "20+" is drop-dead easy to know at a glance.


As for Roll20 not being able to do KAP system, sounds to me that is something to bring up with Roll20 and get them to add another dice resolution mechanic, rather than foisting d20+skill on the rest of us.

Well, that's characterizing what I'm trying to do in the least charitable light possible, but okay. The truth is, the moving crit number and the reversals in dice logic paired with the special exceptions, make it beyond the reach of not only Roll20, but any other digital dice roller I've seen.


Or simply roll d20 like the rest of us peons and calculate the result yourself. :)

Indeed, that is what I do now. But why calculate anything in this day and age? I want to focus on the story and the action, not math.


Best,


M.

But then someone with a modified skill of 30 Sword, would have then same chance to fumble or crit against an opponent with an adjusted score of -5. Doesn't seem right. You're basically saying everyone would have a 10% chance of critting or fumbling in any round. Also doesn't seem right to me. Makes skill count for much less, and randomization count for much more. One could argue that anything can (and does) happen in a fight, but it still seems to penalize extraordinary characters and reward wimpy characters.


I obviously didn't explain myself well. The old rules of skill over 20 would still apply, so you'd get skill-20 (when skill is 21 or over) added to the roll (and straight +skill if the skill is below 0, I think this is already stated in Book of the Manor, Concern my commoners roll, although I don't think it made it to the base rules), and say that:
Result of 1 or below = fumble.
Result of 20 or above = critical.
Sure, this means that you could still fumble on skill 20 (and thus equivalent to skill 19, which still crits on 20) but you could patch that up on a special rule if you really wanted to (I got the impression you wish it to be as simple as possible, even though it might have a small change to the original), or even better, use (skill-19) as the bonus once skill is 20 or more.

This takes care of your -5 modified skill and 30 modified skill having the same chance to fumble or crit: they don't, -5 has 30% chance of fumble and no chance of critting (or succeeding, either, which is a bit harsh*), while 30 has no chance of fumbling and 55% chance of critting.

* Personally, I'd be happier that if your skill is 0 or less, 20 is a normal success. But that would be introducing an exception that you do not like.



But waitasecond. This analysis completely ignore my main issues with the KAP system, which is the additional cognizant load of

1.) Remembering that your crit number can change with every round, depending on modifiers
2.) Remembering several "exceptions to the rule" — those are recounted in my original post — and a couple of reversals in the internal logic of the system.

Is the cognitive load imposed by the one additional calculation more or less than the cognitive load of remembering and applying the pretzel logic described above? That is the key question (to me).


Nope, hasn't been a problem, and as you notice from my above explanation, it would take care of the crit number changing (objection 1 down) and there being no exceptions whatsoever (objection 2 down). Without forcing me and my players to do more math.



Well, you're already comparing roll in opposed resolution anyway, and "20+" is drop-dead easy to know at a glance.


EDIT: My response was for how you suggest handling crits: "If your roll is 20 more than his, you crit."

It is still an additional step which takes a moment. And given that you were complaining about the changing crit number above, this would make it even worse, since it would change on not only your own roll, but your opponent's, too!

I actually did a quick test on a lark with my wife (one of my players), asking her if 37 is more than 20 larger than 15. It took her a few seconds to come up with: "...no? I mean yes! Right?" Granted, she hates math. In any case, her thought process was: "15+20 = 35. 35 is smaller than 37, so 15 is more than 20 smaller than 37, so 37 is more than 20 larger than 15." You can see that is a thought process which is about as long as the whole rest of the opposed rolling mechanism, even with skills of 20+ added to the mix. And you'd have to do that with EVERY SINGLE OPPOSED PAIR. (Even though some of them would be more obvious, like 35 vs. 3.) While this is a sample size of one, it is amply clear that this is NOT conforming to your stated goal (below) of reducing math, but adding it.



Well, that's characterizing what I'm trying to do in the least charitable light possible, but okay. The truth is, the moving crit number and the reversals in dice logic paired with the special exceptions, make it beyond the reach of not only Roll20, but any other digital dice roller I've seen.


I was posting under the assumption that you would like to see 1d20+skill moving forward in official KAP rules. For all the reasons I have already stated, I do not like that idea.



Indeed, that is what I do now. But why calculate anything in this day and age? I want to focus on the story and the action, not math.


Nothing, of course, prevents you from using 1d20+skill in your own game. What I am objecting to is the idea that ADDING more steps and math would make the math less, and very much opposed to making that the official system. Keep it to your houserules. :)

I'd even admit that for a 100% online game with dice rollers, the 2a step I referred to is much less taxing, since you get through that with just typing +skill. I am not familiar with Roll20, so I don't know if it does the comparisons for you, too.

Keep it to your houserules. :)

That's why I posted this in the House Rules forum. :p

I know there's no chance of changing the official rules.


T.

I'd much prefer:
D20 + skill + modifiers: 20+ = success, 40+ is crit, <1 is fumble..
Great thing, is that you CAN also have 30+ as 'partial crit'.. and with opposed high skill rolls you CAN crit if 10+ or 20+ higher than opponent instead
So much more elegant and simpler..
I play with groups of PhD rocket scientists and they have problems grasping old system quickly..


I respectfully disagree with you, my erudite colleague. :) What you have presented is not more elegant and simpler, since it introduces again the interval comparison between opposed skills. Also, you totally change the possibilities of fumbles (impossible to fumble unless your modified skill is 0 or worse) and criticals (impossible to (fully) critical unless your modified skill is 20+, save when it is an opposed roll, since apparently you can critical easier when you are being interfered with???).

Partial criticals are very very easy to add to the current system if you want, simply by adding a confirmation roll on the criticals (roll 1d20 again, and if you succeed in the original modified skill, it is a full critical, otherwise, it is a partial critical). That is actually the system we use (mainly in order to have more differentiation between skills 5 and 15 when they roll a critical). Since this confirmation roll crops up only in about 5% of the rolls, it doesn't slow down the system much at all, although I can see how it might make a dicebot more difficult.



The thing I dislike most old system is that a single +1 or -1 is the difference between crit and fail.. [or even fumble]
Player: I rolled a 20 and I have skill 20, that's a crit!
GM: No, remember that -1 because of your scratched sword... you fumbled...

Addressed in my suggestion by making 20+ result a critical always. :P

That's why I posted this in the House Rules forum. :p


True, true. :)

I went back to BRP to change how the dice work. Crits are much less common at high skill totals but I adopt the 'special' category from BRP as well.

Low roll is always better but a special beats normal and a crit beats special always.

You crit on the tens digit of your skill. If your skill is under 10, then no crits for you. This means that a passion merely improves your chance to crit by 1.
You get a special result on your crit result times 4. If your skill is under 10 you get a special on 1.

All the usual modifiers from KAP still apply so if the ones digit of your skill is over 5 then the normal +/-5 kicks you up to the next crit level and if is under then it knocks you down.

Several surveys have been done, including some in classrooms - the general preference in the US is roll high. About 60:30:10 High:Low:no preference.

I myself collected data with 5th and 6th graders - and I got roughly that same ratio. So Roll Low isn't always better.