• The VOIDRUNNER'S CODEX is coming! Explore new worlds, fight oppressive empires, fend off fearsome aliens, and wield deadly psionics with this comprehensive boxed set expansion for 5E and A5E!

The Rule of 3 - fixing high skill totals

Kerrick

First Post
I got this idea when I was reading Hawken's idea about SR. He proposed changing the check to a simple 1d20+modifiers vs. SR, and I thought... why not apply this to skills as well?

This is a really far-out idea, so keep an open mind for a minute. I'm not sure if it would work myself (I mean, I know it will work, but I don't know if it's practical), which is why I wanted to get some opinions.

I call this the Rule of 3, and it goes something like this: For any skill score, divide the total score (ranks + ability mod + misc. mods) and divide by 3. This is your effective total, which is what you use when making skill checks.

That's the funky part. Here's the part where I prove my point. I'm going to use the existing 3.5 system and the Pathfinder system to illustrate how this works in practice.

Let's say we've got Frank the L1 halfling rogue. With the 3.5 rules, he has a Climb score of 8: 4 (max ranks) + 2 (Str) + 2 (racial). He could make a DC 10check 95% of the time (failure only on a 1), a DC 15 65% of the time (7 or better on the die), and a DC 20 40% of the time (12 or better). He could even make a DC 25 check (overhang or ceiling with handholds but no footholds) a whopping 10% of the time (18 or better on the die) - at level 1. This despite the fact that DC 15 is supposed to be difficult, and just about anything over DC 20 is supposed to range from very hard to nearly impossible for normal (read: non-epic) people, let alone a just-starting-out adventurer.

Under the Pathfinder rules, his score would be 8 also: 1 (max ranks) + 3 (trained skill bonus) + 2 + 2.

Under the Rule of 3, his score would be 2. Suddenly, that DC 15 check looks pretty daunting - he would succeed only 35% of the time! The DC 20 check? Pfft - good luck; he could succeed only 10% of the time. The DC 25 check is impossible, as it should be for a L1 PC.

As you can see, this rule effectively curtails the rampant increase in skill scores - a DC 15 check becomes (and remains) difficult for most PCs under L10, and a DC 20 does the same for those up to L15 or so. Best of all, you don't have to do anything to PCs' or monsters' skills - you use the existing totals, just divide by 3 to get the effective score you use for checks. You would have to reduce DCs, however - it would be nearly impossible (munchkins notwithstanding) to get a check result over 30-40 with this system, unless you're playing very high epic.

And speaking of epic, it would definitely curtail the uberness of epic PCs. No longer will they be making DC 100 checks to swim up a waterfall or dance on a cloud.

This system also means that modifiers are more important - Skill Focus, for example; a simple +1 bonus from bardsong, minor spells, or even aid another become useful again. BTW, I just want to add that I'm considering changing the animal buff spells to the bonuses they add are "end of chain" - that is, they add a +4 bonus to Strength-based checks and skills instead of a +4 to the ability itself. This makes things a lot easier to figure out, and it would add a nice boost when using the Rule of 3.
 
Last edited:

log in or register to remove this ad

Hawken

First Post
I like the idea. The only issue is with the fractions left over. If you divide everything by 3, that means that you divide modifiers by 3, ability score mods by 3, and ranks by 3. That's a lot of leftover if you're working with +1 or 2 mods and ranks non-divisible by 3. However it does get things in line with listed DCs.

What about a different variant to reduce the fractions.

Use the Pathfinder rule for skill ranks (if I'm correct about this): 1 rank per level. Then:
--Reduce the modifiers of ability scores by 1/2: So, a score of 12-15 gives a +1 modifier, 16-19 gives a +2 modifier, etc.
--Racial modifiers are +1; +2 if the bonus is +4 or higher.
--Masterwork bonuses are +1 instead of +2.
--Magic Item/Spell bonuses are +2; +5 if the bonus is normally +10 or higher.

Skill modifier:
Ranks +
Innate modifier (ability score OR racial modifier) +
Misc modiifer (racial* OR Masterwork OR Magic) =
Total Skill Modifier
* Racial can be a Misc modifier if NOT used as an Innate modifier.

Use only 1 modifier for the Innate and one for the Misc; whichever is higher for either category. The only exception would be the Aid Another option. A successful Concentration (my idea) or the same skill assisting with (DC 10) would provide a +1 bonus, +2 if the check is over DC 20. The Aid Another bonus is just that; a bonus and doesn't count as either the Innate or Misc modifiers.

So, a 3rd level Human Rogue with a 16 Cha would have an Diplomacy skill of:
3 (ranks) + 2 (Cha mod) + 2 (Circlet of Persuasion) = +7.

And his ally, a 3rd level Half-Elf Fighter with a 13 Cha would have:
3 (ranks) + 1 (Cha mod) + 1 (racial) = +5 for his modifiers. And if he wanted to Aid the Rogue, he would have to roll a 5 or better on d20 to give his friend a +1 bonus to his Diplomacy check; or a roll of 15 or higher would give a +2 Aid bonus.
 

Kerrick

First Post
The only issue is with the fractions left over. If you divide everything by 3, that means that you divide modifiers by 3, ability score mods by 3, and ranks by 3. That's a lot of leftover if you're working with +1 or 2 mods and ranks non-divisible by 3. However it does get things in line with listed DCs.
I'm dividing the total score by 3, not each individual modifier. At most, you'll have 1 or 2 points left over at any given time. Pathfinder adds a +3 bonus to all class skills, you'll automatically have a +1 in any class skill you have ranks in.

It's the same thing they do with the fractional BAB/saves system - high saves are +0.5; low saves are +0.33. So, a Ftr 10 would have Fort +7, Ref +3.3, Will +3.3. You don't "lose" the remainder - it's just rounded down, but it still counts if you multiclass. For example, the fighter takes a level in rogue, so he'd gain +0.33 Fort, +2 Ref, and +0.33 Will, for a total of Fort +7.33, Ref +5.3, and Will +6.63 (or +7/+5/+6).

Use the Pathfinder rule for skill ranks (if I'm correct about this): 1 rank per level. Then:
--Reduce the modifiers of ability scores by 1/2: So, a score of 12-15 gives a +1 modifier, 16-19 gives a +2 modifier, etc.
You are correct, but I'd rather not mess with ability score modifiers.

So, a 3rd level Human Rogue with a 16 Cha would have an Diplomacy skill of:
3 (ranks) + 2 (Cha mod) + 2 (Circlet of Persuasion) = +7.
Yeah, but we run into the same problem as before - he can make a DC 15 check 60% of the time, a DC 20 35%, and DC 25 10%.

After doing some quick calculations, I think this system might work, but it would definitely need some playtesting. I took a PC with a focused skill (meaning one he'll use a lot, and which works off a high ability score, like Stealth for rogues) and applied: max ranks + stat bonus (+3 to start) + misc mods (2/3 level; this is a rough guess). The results were surprising (I'm only going to list the DC 10 results; subtract 25% for each +5 of the DC to get the other results):

At L1, the PC could make a DC 10 check 65% of the time; 75% at L5, 90% at L10, 100% at L15, and 115% at L20. I'm not sure why the curve flattens out at the higher levels, but a L20 PC has only a 40% chance of making a DC 25 check.

The further out you take it (level vs. DC), the flatter the curve becomes, until it's very hard to make anything above DC 40 (it's a 30% chance for a L45 PC to make a DC 45 check).

So I tried it with skill total divided by 2 instead. It's definitely too high - you're auto-succeeding at a check at the same level (i.e., a L15 is over 100% for a DC 15 check). This leads me to believe that dividing by 3 is the correct way to do it. Although, I could be overestimating the misc. modifiers, too - I checked a few of my PCs, and they didn't have a whole lot of modifiers beyond stat mods.

Finally, I tried with a non-focused class skill (one that is not used as often and doesn't have a high score tied to it, like Climb for rogues). I figured the ranks would be about 3/4 level, reduced the stat bonus, and gave only 1/3 class level for modifiers. The chances are slightly over half what they are for a focused skill - a L20 PC can make a DC 20 check 45% of the time.

Conclusion: As I said, this needs some playtesting to be sure. I find it very interesting how the curve flattens out at the high end, like the law of diminishing returns - you're effectively reaching the pinnacle of what is possible, and it becomes ever harder to accomplish, no matter how high you go. It's vaguely reminiscent of the chase to reach absolute 0 - we've gotten within trillionths of a degree of reaching it, but we'll never actually get there. I like it; it means that you can have a range of what can be reasonably done by mortals. The chances of auto-succeeding on a check don't occur until you're 10 levels over the DC (which is perfectly fine, IMO - that's about how it should be). I'll attach the Excel sheet I used to calculate this, for anyone who's interested.
 

Attachments

  • Rule-of-3.xls
    26 KB · Views: 73
Last edited:

Kerrick

First Post
The more I look at the DCs in the PHB, the more I'm convinced that this is the way to go. A DC 15 check, or even DC 20, becomes meaningless after 5 or 10 levels. The Knowledge skill says that you can set the DC to 20 or 30 for really tough questions... pfft. A moderately min-maxed PC could make DC 30 by L10 with a good roll.

I looked over a few of my old characters that I'd converted, to use them as controls. Jessalie is a L7 rogue; here's what she's got for skills:

Balance +12, Bluff +5, Climb +8, Disable Device +7, Escape Artist +10, Gather Information +5, Hide +10, Jump +10, Listen +9, Move Silently +20, Open Lock +10, Search +7, Sleight of Hand +10, Spot +9, Tumble +14.

(She's not really min-maxed; she just has a few good items.) She could make a DC 15 check with most of her "utility skills" (Climb, DD, Hide, etc.) about 40% of the time, which seems about right to me.

If we use Jess' Phoenix stats...

Acrobatics +14, Climb +19, Disable Device +11, Gather Information +8, Jump +11, Knowledge (local) +8, Perception +11, Search +11 (+14 vs. traps), Sleight of Hand +14, Stealth +14, Use Magic Device +11.

She's doing about the same - her average skill score is +3 to +4, so that DC 15 check is still ~40%. The Pathfinder/Phoenix skill system (they're the same thing, but developed independently) actually grants a boost to most skills because of the +3 class skill bonus, so dividing by 3 would be better in this aspect. I just have to figure out a method to convert 3.5 DCs to Rule of 3 DCs so DMs can do it on the fly.
 

Hawken

First Post
For a standardized DC, how do the numbers work out borrowing from 4e, like this, using the following to demonstrate difficulty of the check:

Common: DC 15
Expert: DC 20
Master: DC 25

At 11th level, increase the DCs by +10.

At 21st level, increase by another +5.

How does that work with your numbers? 4e doesn't encourage contested rolls, but flat DCs--their excuse is that it makes the checks TOO random. But these numbers seem solid and by 11th level when that DC jumps from 15 to 25, there should be little change in the chance of success, reflecting the character's increased skill and acquisition of magical assistance with making the check.
 

Kerrick

First Post
For a standardized DC, how do the numbers work out borrowing from 4e, like this, using the following to demonstrate difficulty of the check:

Common: DC 15
Expert: DC 20
Master: DC 25

At 11th level, increase the DCs by +10.

At 21st level, increase by another +5.
That reminds me of Oblivion - they have all the monsters scale with you, so no matter where you go, you're always facing something of an appropriate level. It's widely disparaged among Elder Scrolls gamers (including me).

How does that work with your numbers? 4e doesn't encourage contested rolls, but flat DCs--their excuse is that it makes the checks TOO random. But these numbers seem solid and by 11th level when that DC jumps from 15 to 25, there should be little change in the chance of success, reflecting the character's increased skill and acquisition of magical assistance with making the check.
Well... judging from my PCs, the end skill score is about half level. So, a L11 would have an average skill score of 5-6; a DC 15 check is 50/50. If we increase the DC to 25, that makes it nearly impossible to make (5%). I don't mind keeping the lower level DCs the same; as a PC gains in power and ability, things that were once difficult become easier until they can be performed without a second thought.

As for the upper level DCs, I'm thinking that somewhere around 30-45, you reduce it by 1/8; 46-60 by 1/4, and after that by 1/3. That way you get DCs that scale more or less evenly up to around 40 (there are a few in the ELH of 90-100+, but we'll ignore those). For things with a set progression (like my alt mastercraft system), I'd have to either reduce the DCs, but that's also fine with me.
 

Remove ads

Top