[3.5] What do All These Numbers Actually MEAN?

[C.Penguin]

Hey there Playground,

Some time ago there was a lovely bit of writing posted on these forums that detailed what being x level meant, and used video game characters for references. I've been looking for this thread but so far I have been unsuccessful and I want to find it because my players recently asked me "what do all these numbers on my character sheet really mean?"

That also got me thinking, has anyone ever read anything that did what the thread I mentioned did but with ability scores instead of levels (as in a 10 strength means this while a 20 strength means this. If no one has seen something along those lines, does anyone have any ideas that could help me with this?

[Zaq]

All roleplaying games are abstractions. D&D 3.5 is no exception. All that 14 INT really means is that you're smarter than someone with 12 INT. How much smarter? Beats me. I don't even know what units you'd use (IQ is a start, but not a finish). STR is the only exception, because it's tied to concrete numbers with the lifting rules. Even so, this way lies madness.

[Prime32]

6th-level is the highest any real-life human being can achieve. Only a handful of lv6 people have existed in history, mostly experts. According to the DMG, a 10th-level character is "legendary".

Every +2 levels is a doubling in power, by the CR system.


Generally you can divide things into four tiers, which people have given various names.
EDIT: http://brilliantgameologists.com/boa...php?topic=9031

[Yora]

Someone did translate several world records into skill checks rolling a natural 20 and on average came up with required skill ranks that could be achieved by 5th or 6th level characters. So as a 7th or 8th level character, a good roll can get you results that are better than anything real humans have ever done.

[Kurald Galain]

6th-level is the highest any real-life human being can achieve.

That's very funny, but it isn't stated anywhere in the rulebooks, is contradicted by real life human feats, and is also contradicted by WOTC themselves.

Because we have legendary people in real life, of course. For example, Alexander the Great. Julius Ceasar. Napoleon Bonaparte. Isaac Newton. Wolfgang Amadeus Mozart. There's quite a long list of these on Wikipedia, and thus by D&D logic they are all in the 12-15 level range.

As a rule of thumb, Olympic-class athletes or Nobel-class scientists require ability scores in the 20-25 range, and level 8 or above.

[Yora]

Interesting statement: Do you have any sources on that?

The one example I mentioned did include all the calculations, though I'm not sure I can find it again.

[PetterTomBos]

How do this change with low op? Or races? I mean, one can do amazing things with little, and boring things with much (plain fighter taking bad feats anyone?)

Is that the good old "pure potential, sadly not refined the way it could be" , or is the power lvl. of lvl.s lower than usual? For example in my campaign, we have char.s lvl. 7, and I dont think they have checks like the godlike check-people.

[Kurald Galain]

So anyway. If you want to see what the numbers mean, a good source is White Wolf. They have all kinds of tables like this:

Strength
* Your character is a wimp and a weakling.
** You're a regular joe.
*** You've exercised a lot, and can trek with a large backpack with no problems.
**** You can easily pull off strongman feats like brick chopping or caber toss.
***** Olympic-class weightlifter.

Of course, we need to map the 1-5 scale to a 3-18 scale. We know that ** is average and that a 10.5 is the D&D average, so a reasonable scale is:
* 3-8
** 9-12
*** 13-16
**** 17-18
***** 19+

[Kurald Galain]

Interesting statement: Do you have any sources on that?

It's easy to calculate. I'll use perform(piano) as an example, but the same principle applies to most other skills.

Joe is an average nobody. He knows what a piano is and can press the keys.
I'm a decent piano player, having had eight years of practice. This means that for me, playing is routine, and Joe does not randomly play better than my routine one time out of twenty.
My friend Peter is a conservatory graduate, and is way better than I am. For him, playing extremely well is routine, and I do not randomly play better than him one time out of twenty.
Johan Sebastian Bach is, obviously, better than Peter.

Now calculate the skill ranks.

• Joe has +0.
• In order to routinely play better than him, I need a +10, so that I can take 10 and end up matching his best random result.
• Peter is better than me. He needs a +20 to match my random results.
• By the same logic, JSB needs a +30 or better. Work out his level from there.

Similar example: Joe is an average lazy guy who doesn't practice much. I go the gym a lot so I have pretty decent stamina; as a result, I can consistently outrun Joe every time. My friend Marsha runs marathons as a hobby, so she will consistently outrun me. And Walter Dix is an olypmic-level athlete and will consistently beat Marsha. The same math applies.

[Spiryt]

As a rule of thumb, Olympic-class athletes or Nobel-class scientists require ability scores in the 20-25 range, and level 8 or above.

Not that it will really work, with 2th level Barbarian with Run and Dash or whatever, you can beat world records in 200m or whatever sprint, all while carrying 50 pounds of stuff with you.

And keep that tempo for a good while.

You can also learn new languages at rather ridiculous rate.

On the other hand, there are plenty fairly simply actions you cannot really take at all without feats, which are rare.

Obviously 3.5 won't ever represent anything very sensibly, but people try anyway.

Personally, I often stat perfectly normal peasants as ~ 5th level, just to have some room to maneuver and differentiate with skill ranks and general numbers.

[sreservoir]

So anyway. If you want to see what the numbers mean, a good source is White Wolf. They have all kinds of tables like this:

Strength
* Your character is a wimp and a weakling.
** You're a regular joe.
*** You've exercised a lot, and can trek with a large backpack with no problems.
**** You can easily pull off strongman feats like brick chopping or caber toss.
***** Olympic-class weightlifter.

Of course, we need to map the 1-5 scale to a 3-18 scale. We know that ** is average and that a 10.5 is the D&D average, so a reasonable scale is:
* 3-8
** 9-12
*** 13-16
**** 17-18
***** 19+

... I'm not entirely sure this is valid in a discussion about D&N 3.5e.

[Tyndmyr]

6th-level is the highest any real-life human being can achieve. Only a handful of lv6 people have existed in history, mostly experts. According to the DMG, a 10th-level character is "legendary".

Every +2 levels is a doubling in power, by the CR system.

This is...very sketchy. It's one possible interpretation. Real life doesn't mesh well with D&D, and trying to represent people in D&D often leads to a mess.

Note that this doesn't generally hold true for player classes, in which many of them do not double in power every two levels. Even top tier classes...I would put my bet with the pair of level 10 wizards against the level 12. Action advantage means a great deal.

[Kurald Galain]

... I'm not entirely sure this is valid in a discussion about D&N 3.5e.

I'm not sure what D&N is

Strength
3-8 Your character is a wimp and a weakling.
9-12 You're a regular joe.
13-16 You've exercised a lot, and can trek with a large backpack with no problems.
17-18 You can easily pull off strongman feats like brick chopping or caber toss.
19+ Olympic-class weightlifter.

Is there anything in particular about this table that doesn't answer the OP's question?

[Mikeavelli]

That's very funny, but it isn't stated anywhere in the rulebooks, is contradicted by real life human feats, and is also contradicted by WOTC themselves.

Because we have legendary people in real life, of course. For example, Alexander the Great. Julius Ceasar. Napoleon Bonaparte. Isaac Newton. Wolfgang Amadeus Mozart. There's quite a long list of these on Wikipedia, and thus by D&D logic they are all in the 12-15 level range.

As a rule of thumb, Olympic-class athletes or Nobel-class scientists require ability scores in the 20-25 range, and level 8 or above.

"Legendary" in this sense they only exist in legends, not real life. People like Odysseus, King Arthur, and Batman.

Refer to Calibrating your expectations for the real world people.

[Spiryt]

17-18 You can easily pull off strongman feats like brick chopping or caber toss.

You don't have to be anything resembling strongman to perform brick chopping though.

It's 'just' learn able trick often accompanying various karate trainings and stuff...

[Kurald Galain]

"Legendary" in this sense they only exist in legends, not real life. People like Odysseus, King Arthur, and Batman.

No, that's not what "legendary" means.

Vlad The Impaler is legendary. He's also a real guy that lived in the 15th century. King Arthur is likely a real historical figure as well. There are plenty of other examples - just google up the phrase "Rock & Roll Legends" for more.

Refer to Calibrating your expectations for the real world people.

It gets old that people keep citing that article, as it is a good example of Critical Research Failure.

You don't have to be anything resembling strongman to perform brick chopping though.

That's a good point. Okay, you need a bunch of skill ranks in Karate. That means that, by my earlier math, you need +10 to that particular skill if you can do it consistently, because Joe Average cannot randomly do it by trying it twenty times.

[Frozen_Feet]

Now calculate the skill ranks.

• Joe has +0.
• In order to routinely play better than him, I need a +10, so that I can take 10 and end up matching his best random result.
• Peter is better than me. He needs a +20 to match my random results.
• By the same logic, JSB needs a +30 or better. Work out his level from there.

I'd argue you only need increments of 5 skill ranks to "routinely" be better at something than someone else, as the DCs typically increase with intervals of 5. This is especially poignant in situations where both parties are allowed to "Take 10".

It's also good to remember that skill checks don't auto-succeed or auto-fail on natural 20s and 10s. So even one rank difference can mean that one party can achieve something the other never will.

Factor in feats, flaws, differences in ability score etc., and the "minimum level" required for several things becomes much lower. For example, to achieve +10 in perform without any other variables, you'd need to be level 7. With Skill Focus and Cha 18, you suddenly reach +11 modifier at level 1 with only four ranks.

[Kurald Galain]

I'd argue you only need increments of 5 skill ranks to "routinely" be better at something than someone else, as the DCs typically increase with intervals of 5. This is especially poignant in situations where both parties are allowed to "Take 10".

Ah, but what if the lower-skilled character isn't taking ten?

I will always, always play better than Joe (and worse than Peter) if we're all taking ten. Since Joe is aware of this, why wouldn't he just roll? If he tries going beyond his routine, thus making a d20 check, he does not suddenly beat me one time out of four, and neither do I play better than Peter 25% of the time.

Yes, feats and other bonuses can apply. But if you say that I have 18 charisma and skill focus, how will you stat out Peter?

[Spiryt]

That's a good point. Okay, you need a bunch of skill ranks in Karate. That means that, by my earlier math, you need +10 to that particular skill if you can do it consistently, because Joe Average cannot randomly do it by trying it twenty times.

Well, this sums up your previous point nicely.

The difference between "Average Joe" and someone with "legendary stat" is like + 5 or +6 to roll?

If someone is 'legendary', and can do something uncanny consistently, Average Joe still can do it by trying often enough...

So, well...

[Kurald Galain]

The difference between "Average Joe" and someone with "legendary stat" is like + 5 or +6 to roll?

I said +30, not +6.

Joe has +0.
I have +10.
Peter has +20.
JSB has +30.

Calculate levels from there. Anything that can't come from stats must come from feats or skill, and hence, from level.

[lesser_minion]

It gets old that people keep citing that article, as it is a good example of Critical Research Failure.

If you honestly think that that article is that bad, would you care to enlighten the rest of us as to why?

[Frozen_Feet]

Similar example: Joe is an average lazy guy who doesn't practice much. I go the gym a lot so I have pretty decent stamina; as a result, I can consistently outrun Joe every time. My friend Marsha runs marathons as a hobby, so she will consistently outrun me. And Walter Dix is an olypmic-level athlete and will consistently beat Marsha. The same math applies.

Within D&D, no, it doesn't. Endurance running would most likely be a function of Constitution check, and some ability checks are explicitly called out as being "bigger score wins, period". Likewise, run speed is affected more by feats and class than by skills, so skill system can't be used as a guideline here.

Ah, but what if the lower-skilled character isn't taking ten?

I will always, always play better than Joe (and worse than Peter) if we're all taking ten. Since Joe is aware of this, why wouldn't he just roll? If he tries going beyond his routine, thus making a d20 check, he does not suddenly beat me one time out of four, and neither do I play better than Peter 25% of the time.

Yes, feats and other bonuses can apply. But if you say that I have 18 charisma and skill focus, how will you stat out Peter?

Ah, but you see, with +5 difference in modifiers, the better party can "Take 10" to always achieve a result the other party can't.

For example, you, with your +5 modifier, will never, ever fail an easy song, even under pressure; while Joe will fail that song 25%. When it comes to average songs, Joe has a 50/50 chance of pulling it off, while yours are 75/25. So under extraordinary conditions you still pull ahead. You can continue this up to Challenging songs, where Joe has 5% chance of pulling it off while you succeed 25% of the time.

However, Tough songs (DC 15) are the most interesting point. You see, here, by Taking 10, you can always play the tough song. Joe can't achieve that without putting himself under extraordinary stress, and even then it's only 25% chance. In real life, the difference between succeeding always and succeeding one-fourth of the time is even more poignant than in the game.

"Joe can randomly beat me some miniscule portion of the time, ergo larger difference is needed" is a poor way of looking into it. I've been shooting for years, but every once in a while some complete newbie comes around and gets the same score (or better!) than I do. This happens even in real life.

Better way to look at it is that you never fail an easy task, and under normal conditions, always complete a tough task. Would you hire a drummer who blows the beat three-quarters a time? Let's look at your example in greater depth.

Joe is an average nobody. He knows what a piano is and can press the keys.
I'm a decent piano player, having had eight years of practice. This means that for me, playing is routine, and Joe does not randomly play better than my routine one time out of twenty.
My friend Peter is a conservatory graduate, and is way better than I am. For him, playing extremely well is routine, and I do not randomly play better than him one time out of twenty.
Johan Sebastian Bach is, obviously, better than Peter.

Joe has Cha 10 and no ranks. Modifier 0.
You have Cha 12 and 4 ranks (Max at 1st level). Alternatively, you have 10 Cha, 2 ranks and Skill Focus (Piano). Modifier +5
Peter has Cha 16, 4 ranks and Skill Focus (Piano). Alternatively he's level 4 with 10 Cha, 7 ranks and Skill Focus. And so on. Modifier +10
Johan Sebastian Bach is level 4, has 20 Charisma *), 7 skill ranks and Skill Focus (Piano). Modifier +15

*) Ability score increase from level 4 and being middle-aged or older.

Results:
Joe can consistently play only Average songs. Under stress, he can fail at even the simplest song.
You can consistenly play Tough songs. You never fails simple songs, even under stress.
Peter can consistently play Challenging songs. He never fails simple or average songs, even under stress.
Bach can consistently play Formidable songs. He never fails simple, average, or tough songs, even under stress.

As you can see, the differences are significant. "Randomness" is a very poor argument against +5 difference being sufficient, as even masters do have bad days. And in normal life, where "Take 10" is standard, the difference is clear-cut.

[Tyndmyr]

If you honestly think that that article is that bad, would you care to enlighten the rest of us as to why?

As just one trivial example regarding the "LOTR can be represented in six levels"....a Balor is a Balrog. The name was changed solely for legal reasons. It literally comes directly from LOTR. A 6th level wizard has no plausible way to kill a balrog. It is also highly implausible that a 6th level wizard, or anyone in D&D would fight this battle for several nights and days consecutively.

Also, IQ is not a complete measure of intelligence, nor is it supposed to be.

[Renchard]

Ah, but what if the lower-skilled character isn't taking ten?

I will always, always play better than Joe (and worse than Peter) if we're all taking ten. Since Joe is aware of this, why wouldn't he just roll? I

Because in real life, Joe can't roll. There just aren't that many places in real life where the result delta a d20 makes is in anyway applicable. Someone who's learned some basic chords on the piano (perform(piano) 1 rank) is never going to put on as good of a performance as some one who is well-trained and actively trying to do well (perform(piano) 10 ranks, taking 10). D&D's skill system may say that it's possible, but it just doesn't happen.

For most skills, anywhere between a d3 or a d6 would be a better die to use to accurately represent the difference between the best results and the worst results that an individual could achieve. There was an old 2e blue book, Creative Campaigning, that advocated such a system, and that I used quite successfully back in the day.

[drakir_nosslin]

That's a good point. Okay, you need a bunch of skill ranks in Karate. That means that, by my earlier math, you need +10 to that particular skill if you can do it consistently, because Joe Average cannot randomly do it by trying it twenty times.

That doesn't work either. I learned how to punch through 30 cm concrete in just a month. Doesn't take 10 ranks of training to pull something off that a untrained person almost never will succeed with.
I'd say that a skill trick is more likely in this case.

And, if we continue on your idea that JSB has 30 ranks in preform, he was lvl 27? That means that he'd survive multiple sword hits to the chest, falling from a sky scraper, consistently outperform people not only in music, but also in numerous other subjects (at least 5 others, if he's an expert) which would vary depending on what skills he chose to focus on. He'd also outfight almost everyone on the planet earth. So, no thanks. I like my experts in the max lvl 6 field.

[Pentachoron]

As just one trivial example regarding the "LOTR can be represented in six levels"....a Balor is a Balrog. The name was changed solely for legal reasons. It literally comes directly from LOTR. A 6th level wizard has no plausible way to kill a balrog. It is also highly implausible that a 6th level wizard, or anyone in D&D would fight this battle for several nights and days consecutively.

Also, IQ is not a complete measure of intelligence, nor is it supposed to be.

I don't know, I think Gandalf could easily have 6 levels of Wizard. The core argument behind Gandalf only being a level 6 Wizard is fair, but I just don't believe he's ECL 6. He's a damn Maiar afterall, gotta be some Racial HD and LA to that...


On topic: I don't really feel like it's really fair to the system to try to force it to model real life. To borrow from another thread on the boards recently, it isn't trying to model real life, it's trying to model fantasy fiction.

And, if we continue on your idea that JSB has 30 ranks in preform, he was lvl 27?

Well his music was pretty...epic.

[Tyndmyr]

I don't know, I think Gandalf could easily have 6 levels of Wizard. The core argument behind Gandalf only being a level 6 Wizard is fair, but I just don't believe he's ECL 6. He's a damn Maiar afterall, gotta be some Racial HD and LA to that...

Pretty much. He's an outsider with near-diefic powers. He routinely stabs people with things. He's more of a gish than a wizard, honestly.

On topic: I don't really feel like it's really fair to the system to try to force it to model real life. To borrow from another thread on the boards recently, it isn't trying to model real life, it's trying to model fantasy fiction.

Well, basically, you eventually come to the conclusion that D&D isn't the best system to model everything. Remember the recent statting of Bruce Willis? It was amusing, sure, but it shouldn't be taken seriously. D&D isn't really meant to do that.

The stats serve gaming purposes, and they don't necessarily match up to real life all that well. That's ok. It's a fantasy world. Killing catgirls is mostly just gonna make your head hurt.

[WalkingTarget]

As just one trivial example regarding the "LOTR can be represented in six levels"....a Balor is a Balrog. The name was changed solely for legal reasons. It literally comes directly from LOTR. A 6th level wizard has no plausible way to kill a balrog. It is also highly implausible that a 6th level wizard, or anyone in D&D would fight this battle for several nights and days consecutively.

Also, IQ is not a complete measure of intelligence, nor is it supposed to be.

Look beyond the arguments about LotR characters, though (looking beyond the fact that a Balor has several abilities that Balrogs are never shown to possess and equating the two is pretty fallacious itself - Balrogs inspired the D&D monster currently known as a Balor, but they're not the same thing).

Look at the breakdown of things like what, say, a level 3 blacksmith is capable of making or how the jump rules (in 3.0) modeled the limits of world record long and high jump fairly accurately in 5 levels.

[lesser_minion]

Also, IQ is not a complete measure of intelligence, nor is it supposed to be.

Which means, at worst, that the article makes a bit of a stretch when it claims that MENSA tries to only take the most intellectually capable 2% of the population.

That's hardly "critical research failure", especially when that particular part of the article doesn't actually have much bearing on the conclusion.

As just one trivial example regarding the "LOTR can be represented in six levels"....a Balor is a Balrog. The name was changed solely for legal reasons. It literally comes directly from LOTR. A 6th level wizard has no plausible way to kill a balrog. It is also highly implausible that a 6th level wizard, or anyone in D&D would fight this battle for several nights and days

Which is interesting, because the article doesn't claim that Gandalf can be statted up as a 6th level wizard. It pegs him at around 10th level.

Things that aren't D&D aren't bound by D&D conventions -- it's entirely reasonable for a character in a book to be mostly comparable to a D&D character of a given level, save for one or two powers that would be either grossly inappropriate or completely nonsensical in D&D terms.

[Frozen_Feet]

As just one trivial example regarding the "LOTR can be represented in six levels"....a Balor is a Balrog. The name was changed solely for legal reasons. It literally comes directly from LOTR. A 6th level wizard has no plausible way to kill a balrog. It is also highly implausible that a 6th level wizard, or anyone in D&D would fight this battle for several nights and days consecutively.

Yes, but is Balor an acccurate representation of Balrog? I don't think so. It possesses several powers that the Balrog in the books just didn't.

I agree with you that Middle-Earth can't be accurately represented by level 6, as Sauron, the Nazguls and several other threats are clearly above that, yet are succesfully warded off by the protagonists. Gandalf, indeed, fights several of the Nazguls on Weathertop, causing flames that can be seen miles away.

That doesn't work either. I learned how to punch through 30 cm concrete in just a month. Doesn't take 10 ranks of training to pull something off that a untrained person almost never will succeed with.
I'd say that a skill trick is more likely in this case.

As I tried to demonstrate before, a difference of 5 is enough to explain feats an average person will almost never succeed in. There's a number of ways to achieve that without needing to get more levels or skill ranks.

It's also good to remember that level 1 represents skill a character has learned during ~20 years of his life. So all their skillpoints didn't get allocated in one go.