Objective and Subjective Ability Values

[Clovenhoof]

Thread inspired by some derailing in the Underpowered Classes thread.

First off, I think that the abilities are essentially all on the same scale, i.e. the relation of, say, Dex 15 to Dex 10 is the same as Int 15 to Int 10. Are you with me there?

Secondly, a value of 10 is average. Most common people will have most abilities at 10.

Now I use the Carrying Capacity table to see what a certain Strength score effectively means. A Str 15 character can carry twice as much as a Str 10 character, so in short, he is twice as strong.

And due to premise #1, the same applies for all abilities - of course this is rather abstract:

10 = average
13 = 50% better
15 = twice as good as average (twice as agile, twice as clever, etc.)
18 = three times as good as average
20 = four times as good as average
and so forth.

Keeping this in mind really helps getting some perspective, and realize how awesome the Heroes really are compared to every Tom, Dick or Harry. A score of 18 isn't just a measly 20% better than 10, as you may be led to believe by the ability modifier, it's three times as good.

Of course, "three times as good" is abstract for many abilities. "Three times as tough" (CON) may mean the guy can keep marching three times as long, but what does "three times as wise" mean? Don't worry, it's not important. Just get a feeling for the relations.

 

[Vortigern]

Tentatively I can express a desire to agree with you in premise however I'm getting an ominous feeling that I'm starting down a path that you know the end of and I don't... :wink:

 

[Clovenhoof]

Well, first and foremost I really just mention this to offer a way for putting the abilities in perspective. I'm afraid that especially Game Masters are prone to forgetting this, so you could say the game world does not appreciate high ability scores the way it should. In essence, every 5 points is a 100% increase.

However, in fact there are other implications, though it may be opening a can of worms -- so please take the following as mere speculation. Just sticking to Strength for a minute, is the steep curve shown by the CC chart actually reflected by the rest of the system? I don't think it is. After a few levels, skill ranks are much more important than the base ability modifier. A character with an ability of 15 has only a marginally (well, 10%) better chance of making any given Ability Check than a normal bloke, when indeed his failure chance should be cut in half, or at least significantly lower. 2 points more or less on a 1-20 range simply don't cut it.

So as I think about it, I'm toying with the idea to make Abilities more significant, which might be achieved by giving _each_ point above and below 10 a 1 point bonus or penalty. So a value of 7 would give -3, and a value of 6 would give +6.

What would be the implications?
* Heroic characters would immediately get much better at all sorts of Skill checks, especially Untrained skills.
* Heroic characters would also get a greater edge at low levels in combat versus average mooks. This advantage would mitigate in higher levels, when Attack and Defense progressions become more important than the ability modifiers, and mooks get waltzed over anyway.
* between high level characters, it would both set the different characters further apart, but still cancel out on similar levels. So a Dex 22 char can dodge the Str 22 char's blow with the same ease as before.
* the biggest difference would probably be Damage, when for instance a Str 22 char wielding a Two-handed weapon get something like 18 points of bonus damage. So this would certainly require some adjustments. Either reduce the Str bonus to damage, or increase the Massive Damage threshold, maybe also increase Hit Points, and possibly Damage Reduction.

As I said, mere speculation, and far be it from me to change the rules unless it's made sure it wouldn't break the game.

 

[Demetrio]

the biggest difference would probably be Damage, when for instance a Str 22 char wielding a Two-handed weapon get something like 18 points of bonus damage. So this would certainly require some adjustments

A possible way round this would be to say the character may divide his strength bonus between 'to hit' and 'damage' as he sees fit. thus with a str +4 bonus one could fo + 1 to hit, +3 dam or +2 to hit +2 dam, etc.

There is still a problem with finesse attacks though and those feats that allow Dex to be used as the modifying stat for melee damage.

Given MD is broken anyway I wouldn't worry too much about that myself.

One thing about making stats more important in this way is that it does devalue skills to an extent as you mention. One may regard that as a good or bad thing, obviously.

On a slightly related note, I quite like the idea of feats like Stealthy giving Dex modifier (obviously depending on the precise feat in question) as the bonus rather than just the flat +2 - it allows the feat to be semi-worthwhile for mid-high level characters.

 

[treeplanter]

He is twice as strong as carrying stuff but not that strong at giving blow. I don't see this as a big deal.

Let say I fight one of those strong guy who push or lift insanly heavy thing, fine he can lift thrice as me but does I really have more chance than me to kill someone with a sword? I don't think the person receiving the blow will even make a difference. Also I'm hundred percent sure I'll climb 4 time better than him.

The thing is in real life training matter a hundred time more than in a heroic fantasy setting, so I think skills represent that well (to a certain extent). In fact I think it would be really stupid that the rather-weak thief who put the max rank in climb is actually a worse climber than the 22 strenght soldiers.

Back on topic, a subjective use of strenght might be the characters with 18 strenght will really impress in a strong-man trial easily reaching the final agains another 18 strenght character. Or when he meet the milicia captain looking for recruit this one is really impressed with the character muscles (and don't care much about his apparence or personnality)

 

[Evil_Trevor]

Me again, hey I'll have to change my sig to "Guess who remembered his password today" .

Anyroad, The d20 dice system sucks, because you roll that clunky die and get a result from 1 to 20 the distribution is even, which is just not how things work in what I like to refer to as "the real world" out here if you have a skill in doing something (say driving) then most of the time you will do it with about the same skill, occasionally you might make a bit of a woopsie, or pull off a nice one, this pattern of results is oft referred to as the bell curve (OK Normal distribution for Math/Stats freaks). To get close to it with a D20 you need to add modifiers to you roll and then set a 'target number', the higher the 'target' the lower the chance of success. However your characters performance still varies wildly from round to round.

If you roll 3d6 (for example) and add the values on each die then you get a close approximation to the bell curve (technically a binomial distribution) You will get results between 3 and 18, but you will only roll 3 or 18 about 1% of the time, you will get a 'average' result '9 - 10' about 60% of the time. A slightly above or below average about 30% of the time (15% above, 15% below). No need to Add modifiers or have a scale of 'DC' checks. Jobs a good one.

My favourite game using this method is the "Free for anyone including Mongoose to use (hint)" FUDGE system. Downloadable from http://www.fudgerpg.com/fudge

 

[Demetrio]

I see.

It's just a different way of doing the same thing actually. Let's say a task has a 50% chance of success.

In fudge you need to roll 11+

In d20 you need to roll 11+

In a % system 51+

If that chance of success drops to 10% then you need to roll about 15+, in d20 19+, in % 91+.

The GM still has to set the correct difficulties to challenge the characters. Also when doing something mundane, like driving, it's unusual to ask for a roll in an rpg. It's the stressful and extraordinary that require rolls. The 'determine chance' and roll works fine for that. And it works with any system.

 

[Clovenhoof]

Yep, I agree.
To add one more to the list:
in Savage Worlds, you roll one die of variable size, and the target number is (almost) always 4. The average person has attributes of d4, so his chance of success is 25%. If you have a d6 attribute, your chance is 50%, and so on. Special training or other circumstances can increase your result by +2, so your chance is 50% with a d4, 83% with a d6, and so on (essentially, you only fail on a 1).

Anyway, back to D20.

If you have an ability of 13, which according to my CC explication is 50% better than average, your chance to beat a DC 11 is just 5% better than normal, i.e. you have to roll a 10+ instead of 11+. This does not feel like a Good ability value at all!
The trouble with a linear distribution is that the picture looks completely different for other initial chances than 50%. If the DC is 20, that's a 5% chance for the average "10! guy, but a 25% chance if you get a +5 to the check, so that's about five times as good.

A step towards a solution might really be a bell distribution, such as achieved by rolling 3d6 instead of 1d20. That just reminds me I had planned to try exactly that, at least for non-combat checks, in my game but so far I haven't got around to it.

 

[Ichabod]

I hate a d20 because it's too much variance. But, I also hate 3d6 because you get boring as hell rolls without the extremes that make things more interesting. d10 is okay but would require reworking massively d20. I'd be inclined toward 2d10 to get something close to d20.

Anyway, I'm not clear on what the point of this thread is. So, something is flavorwise double something else? And? In Hero/Champions, every 5 STR doubles lifting ability to where you can't realistically simulate Marvel superheroes and you get silly stuff like an 85 STRer is twice as strong as an 80 STRer. The 6% increase in damage might work flavorwise, but the idea that 5 character points makes someone twice as strong is absurd.

I thought the point of the thread was going to relate to the idea that some attributes are easy to measure mechanically but have wildly differing benefits depending upon how the game is run while others are pretty much the same from campaign to campaign. I don't know that attributes really work this way. Maybe some GMs care that someone's CHA is 22 vs. 19 vs. 12 in terms of reactions, but I'd think it would have more to do with skill checks than base attributes.

Of course, there are characteristics that matter a lot more for different characters in different campaigns, but why does that need commenting on?

I'm much more confident that classes, races, feats, skills, sorcery, spells all have this campaign-dependent feature. Just take for instance someone's reaction to my mentioning Decipher Script in the blindness thread or how Vincent doesn't seem to think K: Religion and other KSs are useful (or his advice in H: Fiercest about where to put skill ranks which would be utterly suicidal in our campaign). These are extremely important skills in our campaigns for knowing what the hell is going on, i.e. for knowing how to kill the bad stuff. Makes any crosscampaign analysis fairly timewasting in the grand scheme of things, though it is interesting to see how people don't agree on what problems exist.

 

[treeplanter]

Wow I must admint I don't quite get where you guy want to go with the normal distribution. Normal distribution is pretty easy to attain in d20 just use the take 10 more often you'll get your nice gaussian curve!

Clovenhoof I don't follow your logic... Basicly you say that it suck that somebody have 13 strenght he is just 5% better than the guy with 11 strenght?

The way I see it let say i'm trained in swim, so I've 5 rank in it, so I've roughfly 75% odds to beat a DC 11 swim check. the guy with no training but who is a bit stronger (str 13) will be slightly better than the untrained guy but lot worse than the trained guy.

On the other hand the brute with 18 strenght will be almost as good as the trained guy, but if the last one train more i will totally eclipse the brute wich is again pretty logical.

If it is DC 20 it is normal that the guy with no training have no real chance of success. 5% almost seem generous to me as Impossible would be appropriate. If you roll 3d6 (but DC is 18) it is even more marked: The guy with the +5 will success roughly with the same % but the guy with no training will be at 216 to 1...

In all case I don't see any of this going anywhere!?!?

 

[treeplanter]

I guess in the end Ichabod is right

 

[seanbickford]

Like the topic.

My take on ability scores is that they should progress linearally in effect, but be distributed like a bell curve. In other words, the skill modifiers should be +2 (+1 would make more sense) per step up the ladder, but 3's and 18's should be extremely rare, constituting only a very small % of the population. Beyond that--19+-- should be even more rare.

Yeah, rolling a d20 is just too random, and I agree with Ichabod that 3d6 seems too mundane (though I've never played with 3d6 a lot).

Back to the original point, I think the strength tables are just wacked and should be brought into line with a linear system. I think having a system that has modifiers on some sort of curve would be needlessly confusing. Just put character progression on a curve, but leave the scale linear.

 

[Clovenhoof]

I'd be inclined toward 2d10 to get something close to d20.

That's another idea I've been considering. Didn't we have a thread about that last year? 2d10 might actually work quite well as a compromise between fully linear and very bell-like distribution.
Question is, would you just apply this to skill checks, or also in combat?

Anyway, I'm not clear on what the point of this thread is. So, something is flavorwise double something else? And?

I don't know if you've ever experienced it in your games, but sometimes players (including the GM) would say something like a 14 isn't very special (since it's just 10% on a d20), and I think that's a pity.

Back to the original point, I think the strength tables are just wacked and should be brought into line with a linear system.

Well, that's of course another possibility; in fact most games handle it that way as far as I'm informed. You could say, for instance, that every point of Str gives you 4 pounds of CC without encumbrance, and that a Str 20 is simply twice as strong as a Str 10.
(Actually, I think those D&D CC tables were designed with loot-hauling in mind.)
However, I think that would be a bit boring, or in fact inflationary, as there would have to be quite a lot of blokes with Str >20 in that setup.

Don't get me wrong, I'm in no mood to redesign the whole system, but some tweaking here and there might not be the worst thing to do.

 

[treeplanter]

Just by curiosity did you even bother using the carrying/pushing/dragging capacity?

 

[Clovenhoof]

Why? It has nothing to do with this matter. The higher load tables scale the same way as the light load one.

 

[Teriudin]

I In Hero/Champions, every 5 STR doubles lifting ability

Doesn't the Conan rulebook (and thus D20 in general) also double lifting ability every 10 points? (or is it quadruple?) I think it is clear on the 8's (8,18, 28, etc)

Not that this contributes in any way to this thread, just though I'd point it out.

I would like more "objective" ways to compare ability scores in the same way STR has such a clear delineation.

For example:
First, an easy one
INT: Multiply by 10 to get a rough idea of the character's IQ

Now, a not so easy one:
CHA: Measure's your physical attractiveness. In order for you to be noted as "attractive" to another party, subtract the CHA modifier from the high to the low. That is how many drinks are required for the more attractive person to feel that the other person is physically attractive. Ex. Joe the Ugly has CHA 6. Sexy Sue has CHA 14, but has had 5 drinks and is well past sloshed. Their difference in CHA modifiers is 4, which is less than the state of drunkeness Sue is in. Thus, she regards Joe as attractive and who knows, maybe Joe has enough diplomacy, bluff or even intimidate ranks to pull something off.

Yes, I just used a Beer Goggles scale to define CHA. Sue me!

 

[Clovenhoof]

Alcohol - the Potion that you drink to increase another person's Charisma. ^^

As for Strength: yes, in D20-based systems Strength is doubled every 5 points. Difference between Conan and Champions is that in Conan the absolute maximum for humans is 29, and there's no way to get even close to regions like 80 or 90.

 

[Clovenhoof]

*bump* with a new aspect to the topic:
comparing the value of each Ability to each other. Of course it always depends on what kind of character you're playing, but even to achieve one desired effect there are different ways to skin a boar.

For example: Survivability.
There are three ways to survive enemy attacks: Parry (Strength), Dodge (Dex) and Soak (Con). I used to be very Dex-oriented, figuring that the best way to survive is always not getting hit. But just the other day simple arithmetics made me reconsider:

Defense-Wise, Str and Dex each have one primary use - each point of Modifier increases your Defense by 1, which on a D20 corresponds to 5%. So each 2 Ability points increase your lifespan by 5%. Not bad.

Con, on the other hand, gives you 1 HP per Level (for the first 10 levels), and considering that Hit Dice go from d6 to d10, and that the maximum HP gain from your HD is 10 points per level, this means that each 2 points of Con increase your life force by at least 10%.
If you actually roll your HP the cold hard way and have an average gain of HD/2, the increase by Con is even better, about 20% per 2 points.

Each ability also has secondary uses. Dex also increases your Initiative so you get a slightly better chance of striking before getting struck. Str increases your offensive power so you get a better chance of killing before getting killed. But Con increases your Fort, which is the only way to survive Massive Damage when the shit hits the fan.

Looks like I've seriously underestimated Con in the past. I'll make up on that with my new character. ^^

 

[Ichabod]

Well, there's also things like armor and magic.

My observation of Con over levels for a fighter is that the main benefits relate to recovery. Going from dying to not dying is a bit better when you squeeze out a few more HP. Ability damage, even of the Con sort (the most annoying sort), fixes reasonably quickly. Saves are just so random.

For a non-fighter, the HP impact and save impact are relatively more impactful.

If you want an example of why I can't get all that excited by Con, there was one session where we were comparing HP totals. One character was down 30-something. Another was down 50-something. I mentioned how I was down 90-something. Having a 30 full-HP edge on another character really didn't mean anything. What precipitated this was being hit by a trap while at full HP and failing the MDS (which happens), so one effect took away over 90 HP. Sure, the recovery was marginally better, though there was only the short term care possibility, so the next two sessions were spent limping around waiting for Diehard to trigger again.

While (going back to another thread) every character can pretty much have a dump stat, no attribute is useless. For pure mechanical benefit, having everything high is great. Still, it's not so much Con being underrated that I would embrace as much as Dex being overrated and Str being broken.

There's also some interesting intangible bits. I find that GMs, myself included, like to take down the most threatening combat characters. That means going after STRers. My high DEX, high CON character is not remotely threatening in combat so is easily ignored. So, it would make sense that making your character less dangerous makes your character less of a target, but then, your character is less of a target because your character is less dangerous. I think it's better to just eviscerate everything within reach when worrying about combat ability and not get overly fancy.

 

[treeplanter]

Clovenhoff you can't compare 1 point of defense as a 5% increase in survival with a 5% hp increase the same way cause defense work on each attack, while your hit point go down in one hit.

Just simulate:

A fifth level character have 50 hp and hit 10 dmg 50% of the time versus a character have 55 hp (10% more hp) and hit 10 dmg 45% of the time: the first would last 11.1111 round while the second will last 11 round.

But if the second character have 60 hp (20% more) then he will las more (12 round vs 11.111 round)