A fundamental mechanical problem I have with d20/D&D/PF

[VanIsleKnight]

DISCLAIMER: Before I begin I would like to emphasize that I'm bad at math. This post will reflect that, and I encourage my mathematical betters to create better examples. For these examples I'm pretending 1's and 20's in a stat contest isn't an auto-succeed or failure. They're just values of 1 and 20.

I've come to dislike the 'swingy' nature of the game when it comes to d20 rolls for determining success or failure of any given thing. Discounting the GMs who treat every 1 and 20 as an auto-success or fail regardless of what the roll is for, there have been too many instances where the dice have landed a certain way and everyone at the table can't help but scratch their heads and go "huh?" as we try to figure out how the dice rolls translate into the story.

Spoiler: Example 1

Show

Let's take two humans and have them arm wrestle (as an example, any basic test of Strength/Stat will do) a third human.

-Average Joe: Str 10 (+0)
-Mister Hiro: Str 18 (+4) (human 'maximum' despite without magic or buffs you can achieve a score of 25 at level 20)
-Frankie the Control: Str 10 (+0)

Joe and Frankie are equal in terms of strength, they're also not particularly weak or particularly powerful. They're very average, and have a 50% chance of beating one another in the arm wrestle.

Hiro though is the strongest a normal human being could 'ever' be, but he's only 20% stronger than Frankie is. I'm assuming that means that 70% (7 arm wrestles out of 10) of the time, the strongest human being is going to beat the average human being in an arm wrestle. What?

Interested in pursuing science no matter the cost, I myself am going to arm wrestle Hiro at 15 years of age and as an adult in the next example.

Spoiler: Example 2

Show

Now I myself am not a particularly strong individual. I would venture my own personal STR stat to be around 9. My younger self easily a 6 or 7.

Adult me has a 25% chance of beating Hiro (again, the strongest human being a normal human could possibly be) in an arm wrestle.

Teenager me has a 20% chance of beating Hiro.

Now you could make up reasons why Hiro lost those matches: he got cocky, he felt like giving a free win, he suddenly lost focus, whatever. We as players and GMs are expected to do that, and we do. And again it doesn't have to be an arm wrestle, it could be any basic contest pitting one stat against another stat. Still, the number of times Hiro loses doesn't add up to me, he should be winning almost 100% of the time, especially against teenagers or even adolescents.

Yes, you can do arm wrestles differently instead of a single die roll, that isn't the point of this post, hence why I pointed out repeatedly you can swap it out for any other opposing stat check. Bet someone is still going to mention running arm wrestles as 3 opposing checks as opposed to one. >: D

[icefractal]

That's why there aren't a lot of opposed stat checks in D&D. They're very swingy, they ignore level, and as a result they yield a lot of strange results. So ... don't use them. Personally, I would do something like arm wrestling with a smaller and more curved RNG - 3d4, perhaps.

For skills, the difference is generally more significant, although even then, I think that rolling the standard way only works for circumstances that are somewhat chaotic. In a controlled contest, perhaps both sides taking 10 would be appropriate, or rolling off X times and seeing who won more often.

A quick hack, for cases where you don't want to make it a lengthy contest, and want to keep things similar to standard, is for both sides to roll 3d20 and take the middle one. Sort of a middle-ground between rolling and taking 10, it could be appropriate for semi-threatened situations.

[shadow_archmagi]

Yeaaaaaaaaaaah d20 kinda sucks like that.

[VanIsleKnight]

It isn't just stat checks, but most d20 rolls in general. I've found the game very swingy in general; it's enjoyable enough to serve as entertainment but I find the system itself deeply flawed and clunky. The fact that it needs 'hacks' at all does not help it's case in that regard.

Skills can be substituted for the examples as well, as well as basic combat.

I understand that there are in-game modifiers as well as circumstance bonuses made up by the GM, but that doesn't elegantly or sometimes even adequately satisfies the problem.

[CarpeGuitarrem]

Yeah, it's a commonly-known problem in d20 games. The best fix I'm aware of is to replace the d20 with 3d6 for all of your rolls. Almost the same range of results, and the average roll is the same.

[Airk]

That's why there aren't a lot of opposed stat checks in D&D. They're very swingy, they ignore level, and as a result they yield a lot of strange results. So ... don't use them. Personally, I would do something like arm wrestling with a smaller and more curved RNG - 3d4, perhaps.

For skills, the difference is generally more significant, although even then, I think that rolling the standard way only works for circumstances that are somewhat chaotic. In a controlled contest, perhaps both sides taking 10 would be appropriate, or rolling off X times and seeing who won more often.

A quick hack, for cases where you don't want to make it a lengthy contest, and want to keep things similar to standard, is for both sides to roll 3d20 and take the middle one. Sort of a middle-ground between rolling and taking 10, it could be appropriate for semi-threatened situations.

Them's some wacky houserules you have there, and should not be considered for any discussion of how "d20 systems" work.

So yes. d20 games have two big issues.

#1: The d20 is a flat, very large spread of numbers. A 1 is just as likely as a 10 which is just as likely as a 20.
#2: The spread of character bonuses is not very large compared to the spread of random numbers. A "large" difference in ability in most d20 games is like +10. That means when the person with +15 rolls a 5 and the person with +5 rolls a 16, the person with the +5 does better.

These two factors combine to create very swingy results. These sorts of things are counteracted by other game systems in two ways:

#1: Use multiple dice. This creates a bell curve. Simply substituting 3d6 for d20 means that situations in which a person with a +15 losing to a person with +5 become MUCH rarer.
#2: Make the 'ability spread' wider. You see this in a lot of percent based games. If "average" is +0, and "really good" is +90% or more, you'll see the +90% person losing less often.

But basically, if this annoys you, don't complain about it. Go play something else. There's an infinite number of games out there, and basically none of them use d20s for everything except for D&D, it's "d20" derivatives, and Pathfinder, which is basically a d20 derivative.

Even better, in your particular case, you have a perfectly clear argument for why you don't like D&D: You don't like games with swingy results. So find your own game and enjoy it.

[VanIsleKnight]

But basically, if this annoys you, don't complain about it.

Translation: Never criticize anything ever, and go away you silly king-type!

Currently the best I've found is FATE, we attempted to do GURPS once but that was... yeah. I don't know anybody who plays GURPS and I respect their ability to do so for being able to use something so expansive.

We'll still play Pathfinder since it's still enjoyable enough, however I was unaware of the 3d6 variant rule and will probably give that a shot.

But seriously, never say "Don't complain, go play something else" about any system. Every system that warrants criticism ought to be criticized, else people (specifically game designers) won't think there's anything wrong with it.

[TheCountAlucard]

Actually if neither randomness nor skill is a factor, a roll generally isn't called for; the guy with the higher Strength (after all modifiers) wins.

You gotta remember, you only call for a roll if there's a reasonable chance of things not going right (and if the results would have any sort of meaningful impact on the game). I don't have the 20th level PCs roll out attacks or initiative if they want to kill an ordinary rat, and I don't have the PCs roll opposed Intelligence checks or whatever if they're just playing checkers in the bar to kill time.

[Tengu_temp]

This is why I prefer to roll 2d10 instead of 1d20 (I don't play DND, but I do play M&M) and set the standard crit range to 18-20 to compensate. Suddenly it becomes much harder to beat someone with a higher bonus than you, and even a difference of 2-3 points becomes much more visible!

[Pex]

It's really a matter of taste. It's just something you don't like. However, if you're still going to play D&D and have to deal with the d20 you'll need to either lump it (being sincere, not snarky) or find a solution.

One option it to use an idea 3E's Unearthed Arcana introduced and 4E has as a staple - multiple d20 rolls. For a given task you roll Xd20 and need to get Y successes before Z failures. For opposed rolls it's whoever gets to Y wins first wins. The problem is determining the values for X, Y, and Z. It depends on the task that needs the roll and how easy or hard it is. It also takes up real world time to roll Xd20 and figure out the success or failure. Multiple instances of time for Xd20 rolls adds up over a game session.

Another option is to use the Advantage/Disadvantage system 5E introduced. When you have Advantage in a task roll 2d20 and use the higher roll. Disadvantage is roll 2d20 and use the lower roll. It's possible for one person to have Advantage while at the same time his opponent has Disadvantage. You can make an inquiry in the 5E forums for more information about the system, its pros and cons.

[Airk]

Translation: Never criticize anything ever, and go away you silly king-type!

You need to get your translator checked, Mister, because what I ACTUALLY said was "There are tons of awesome games that aren't D&D, so if you don't like the d20, F- that game, go play something better!" Only I was trying not to talk down to people who really enjoy d20.

But seriously, never say "Don't complain, go play something else" about any system. Every system that warrants criticism ought to be criticized, else people (specifically game designers) won't think there's anything wrong with it.

Because, you know, I totally haven't already analyzed the crap out of this problem or something, in the same post you are whining about. Nor has anyone else, ever, EVER examined this problem. WTF, sir?

I think "If you don't like it, play something else." is EXCELLENT ADVICE in the RPG community, because SO MANY PEOPLE are all "I am playing D&D, and X, Y and Z are really bothering me, how do I fix this? It's ruining my game!" when the answer is "You don't have to, tons of excellent game designers have already fixed it, and you just need to pick a game that's better designed than the ol' warhorse."

Or, to put it another way, the RPG community needs fewer people playing D&D because it's "what they know" and more people making informed decisions about what games they are playing. So if you have a good reason to not play D&D, you should NOT PLAY IT. To do otherwise is like saying "I really like soccer, but I wish I could use my hands and throw the ball into a basket" and then insisting on continuing to play soccer.

[TuggyNE]

One way to deal with this is to use 2d10, possibly with Next-ish dis/advantage.

Another way is to have the d20 only govern part of the result, like you do in combat. If you succeed on one check, it just means you've made d4/d6/d8/whatever progress toward the final goal, so usually multiple checks are required to get through the whole thing.

[Grinner]

Or, to put it another way, the RPG community needs fewer people playing D&D because it's "what they know" and more people making informed decisions about what games they are playing. So if you have a good reason to not play D&D, you should NOT PLAY IT. To do otherwise is like saying "I really like soccer, but I wish I could use my hands and throw the ball into a basket" and then insisting on continuing to play soccer.

While I agree, this position has a number of logistical problems.

First, there's the issue of actually obtaining the books. You can make do with just one, assuming you're not playing online, but it's kinda nice when everyone has their own copy. If you are playing online, well, good luck.

Second, each of these games tend to represent a significant investment of time. I once thought about buying the Game of Thrones RPG, but one look at the table of contents had me reconsidering that notion. People spend a lot of time learning to play these games and becoming competent enough to go play without interruption takes even more time. Here there is a case to be made for rules-light games, but some don't find those quite as satisfying.

This is an expensive pastime, in terms of both time and money. Metaphorically speaking, having a common language is not without benefits. It'd be nice if that common language were a little better designed, but that's the baggage it carries.

[Airk]

While I agree, this position has a number of logistical problems.

First, there's the issue of actually obtaining the books. You can make do with just one, assuming you're not playing online, but it's kinda nice when everyone has their own copy. If you are playing online, well, good luck.

What is this? 1994? Getting books and PDFs has never been easier.

Second, each of these games tend to represent a significant investment of time. I once thought about buying the Game of Thrones RPG, but one look at the table of contents had me reconsidering that notion. People spend a lot of time learning to play these games and becoming competent enough to go play without interruption takes even more time. Here there is a case to be made for rules-light games, but some don't find those quite as satisfying.

Good games also don't years to learn. You can pick up enough about Dungeon World to be able to play it in 15 minutes. Running it? I dunno, it took me maayyyyybe 6 hours to read the book, and I feel entirely mechanically prepared at this point. I don't think it took me much longer than that to learn The One Ring.

The more games you know, the easier it is to pick up new ones. If all you've ever played is D&D, then it may take longer, but "It's going to take time!" is no more of an excuse in RPGs than it is in the "I'd rather be playing basketball" example I used earlier. Do you want to be doing something you enjoy, or not? I tend to find things that I enjoy worth taking a couple of hours to sort out.

A couple years ago, we were starting up a new game, and we just kinda said "We'll run D&D because we know it". It was okay. Then more recently, we started to say "You know what, D&D isn't good at this." and started exploring other games, and our gaming has been much better for it.

The biggest time investment in games is playing them, not learning them.

This is an expensive pastime, in terms of both time and money. Metaphorically speaking, having a common language is not without benefits. It'd be nice if that common language were a little better designed, but that's the baggage it carries.

No, actually, RPGs are REALLY CHEAP these days. Monetarily at least. Time-wise, they are certainly no more time consuming than any of a number of other hobbies. You could learn and play an entire campaign in a game system in the time a lot of people sunk into Skyrim. If you can spend the time to play D&D, you can spend a couple extra dollars and hours to play something else.

Not sure what your language metaphor was about at all, either.

[Grinner]

What is this? 1994? Getting books and PDFs has never been easier.



Good games also don't years to learn. You can pick up enough about Dungeon World to be able to play it in 15 minutes. Running it? I dunno, it took me maayyyyybe 6 hours to read the book, and I feel entirely mechanically prepared at this point. I don't think it took me much longer than that to learn The One Ring.

The more games you know, the easier it is to pick up new ones. If all you've ever played is D&D, then it may take longer, but "It's going to take time!" is no more of an excuse in RPGs than it is in the "I'd rather be playing basketball" example I used earlier. Do you want to be doing something you enjoy, or not? I tend to find things that I enjoy worth taking a couple of hours to sort out.

A couple years ago, we were starting up a new game, and we just kinda said "We'll run D&D because we know it". It was okay. Then more recently, we started to say "You know what, D&D isn't good at this." and started exploring other games, and our gaming has been much better for it.

The biggest time investment in games is playing them, not learning them.



No, actually, RPGs are REALLY CHEAP these days. Monetarily at least. Time-wise, they are certainly no more time consuming than any of a number of other hobbies. You could learn and play an entire campaign in a game system in the time a lot of people sunk into Skyrim. If you can spend the time to play D&D, you can spend a couple extra dollars and hours to play something else.

Not sure what your language metaphor was about at all, either.

From your perspective, these things may be true.

Personally, I have trouble justifying $20 for a single book, and I'm certainly not accustomed to paying full retail. I suppose getting PDFs with the PirateBay discount is an option, but that's not one I prefer.

Personally, getting players together on a different system is a chore. Homebrewing rules is also a chore, but less so.

Your assertions may be true for you, but they're certainly not true for me.

[Jay R]

Don't use an opposed strength roll on anything unless it's random, and the weaker person can reasonably win.

In an arm-wrestling match or a tug-of-war, for STR 10 vs STR 18, I wouldn't even roll.

The problem is not the mechanic you're using to generate the random number, but the fact of generating a random number at all when the situation is not random.

[Tarqiup Inua]

I can't speak for the system, but you are making few assumptions, here.

Physically strong human (such as harbour worker used to carrying great burden) and good wrestler aren't totally the same. You can defeat strong man with little stamina by one well aimed kick in the groin or by forcing your fingers in his eye socket (equivalent of one unarmed attack).
Good wrestler would have higher constitution and perhaps better dexterity as well. He might also have maxed hit dice for first level (we are speaking of a professional).
He will definitely have a feat for fighting unarmed.

Mind these matches usually have special rules - the task is seldom punching someone to unconsciousness - many sports and many types of ocmpetitions even forbid that.
If you want your match to mirror the reality, go for something different.

For example - if the sport is anything like greco-roman wrestling, you may decide the match is solely based on grappling without dealing damage. The ways to win - holding someone pinned for over 1 round or dragging someone all the way to the edge of the circle and throwing them out, require more than just one grapple check and will favour the stronger opponent.

If you want your match to resemble actual fighting, you are still not out of options. The rules, again, may forbid using moves that can cause serious harm, which could translate into, say, low damage reduction with scaling based on endurance. The advantage goes to the strong hitters, again, because they can overcome it.

I could imagine allowing a bluff check, where - if you succeed, you can hit your opponent using illegal move without the referee noticing, and ignore your foes' damage reduction 3/dirty fighting.

It takes some more work to make it sensible, I guess...

[Rhynn]

Have to correct the numbers...

Spoiler: Math

Show

At equal bonuses, you're going to have 5% draws and 47.5% victories for both.

At +0 vs. +4 (a 4-point difference), it's 4% draws, 30% for the lower bonus, and 66% for the higher bonus.

At -1 vs. +4, it's 3.75% draws, 26.25% lower, 70% higher.

At -2 vs. +4, it's 3.5% draws, 22.75% lower, 73.75% higher.

You basically forgot the draws, that's all.

For fun, +5 vs. +15 is 2.5% draws, 11.25% lower, 86.25% higher.

The solutions have been mentioned; multiple dice tend towards averages, etc.

For instance, if you use 3d6 instead...

Spoiler: Math again

Show

At +0 vs. +4, the odds that the +0 beats the +4 are 14.46%.

At -1 vs. +4, the odds are 9.65% for the weaker.

At -2 vs. +4, the odds are 6.08% for the weaker.

Those odds are lower, but they still look too high, don't they?

For funsies, +5 vs. +15 is 0.99% odds for the lower bonus.

I think Jay R is spot on here, ultimately: the only way you can model this sort of thing well is to not even roll for the average guy against the strongest guy. I really doubt that, over a statistically significant number of tests, we'd see an average guy win at arm-wrestling against a strongest guy even 1% of the time.

It comes down to what you want from the system. D&D 3E is not supposed to model realistic results in arm-wrestling matches with opposed Strength checks; that's not any kind of priority for the system.

Also, Airk is right about choosing better systems for yourself. Excuses are excuses; there's plenty of legally free RPGs, and plenty of RPGs worth money and time. (Also, if you can't spend $20 per book, how the heck can you buy D&D?)


Edit:

one well aimed kick in the groin

fingers in his eye socket (equivalent of one unarmed attack).

punching someone to unconsciousness

Dude, the example was arm wrestling. You either missed the whole point of the OP, or I really feel sorry for anyone who's ever arm-wrestled you.

[Tarqiup Inua]

Dude, the example was arm wrestling. You either missed the whole point of the OP, or I really feel sorry for anyone who's ever arm-wrestled you.

Or perhaps it is you who missed the whole rest of my post where I happened to be talking exactly about that - rules in fighting competitions and ways to introduce them?

I was merely pointing out that unarmed attack is a very bad representation of wrestling.

[CarpeGuitarrem]

All hail the d20, sower of discord in this thread.

[Rhynn]

I was merely pointing out that unarmed attack is a very bad representation of wrestling.

But you were the one who brought up unarmed attacks, the OP's example was opposed Strength checks... and the thread is about the mathematical results that the d20 system produces with opposed checks, not about unarmed combat... so your post was kind of a giant non-sequitur.

[Tarqiup Inua]

But you were the one who brought up unarmed attacks, the OP's example was opposed Strength checks... and the thread is about the mathematical results that the d20 system produces with opposed checks, not about unarmed combat... so your post was kind of a giant non-sequitur.

Others said multiple d20 checks can resolve the problem, I provided a way to make a wrestling match into several grapple checks (which happen to be opposed checks), and pointed out how to make it work with or without unarmed attacks, so that seems relevant to me.

I agree d20 has its oddities. You'll get no argument from me, there...

[1337 b4k4]

From your perspective, these things may be true.

Personally, I have trouble justifying $20 for a single book, and I'm certainly not accustomed to paying full retail. I suppose getting PDFs with the PirateBay discount is an option, but that's not one I prefer.

Personally, getting players together on a different system is a chore. Homebrewing rules is also a chore, but less so.

Your assertions may be true for you, but they're certainly not true for me.

I think you're missing the point that these days there are more free rule systems than you can shake a stick at. If you want D&D, there's the entire retro-clone industry, then there's Dungeon World, Basic Fantasy, Eclipse Phase, Stars Without Number, everything here, the last few years around the holidays Classic Traveller has been free on RPGNow.com, then there's the entire free and pay what you want section on RPGNow.com (which includes all the old WEG d6 systems). In this day and age, if you have a computer or an ebook reader that can display PDFs (or access to someone who will print it out for you) you could game for your entire life playing a new system entirely every campaign, without ever paying one cent for a rule book and do it all perfectly legal.

Now, getting other players might be an issue, but if your players are as frustrated as you are with the failings of a given system, then getting them to switch shouldn't be an issue with the free games because it requires no outlays.


Edit
----------------
As to the OP

Now you could make up reasons why Hiro lost those matches: he got cocky, he felt like giving a free win, he suddenly lost focus, whatever. We as players and GMs are expected to do that, and we do. And again it doesn't have to be an arm wrestle, it could be any basic contest pitting one stat against another stat. Still, the number of times Hiro loses doesn't add up to me, he should be winning almost 100% of the time, especially against teenagers or even adolescents.

One thing you could try is to bring in the idea that failure on a contested roll doesn't necessarily mean failure of the immediate action but it means there are consequences for the action. For example, your arm wrestler might still win the match, but maybe got a little carried away and broke their opponent's arm in the process. Turns out their opponent was the son of a petty lord looking to score political points...

Or maybe the local guard shows up at the tavern to bust up the gambling, turns out arm wrestling is considered gabling in this town and now the players are under arrest...

Obviously stuff like this requires a lot of trust in your DM to be fair, and isn't applicable to every situation, but it can certainly help turn the swinginess of the game into a positive instead of a negative.

[Mando Knight]

From your perspective, these things may be true.

Personally, I have trouble justifying $20 for a single book, and I'm certainly not accustomed to paying full retail. I suppose getting PDFs with the PirateBay discount is an option, but that's not one I prefer.

Personally, getting players together on a different system is a chore. Homebrewing rules is also a chore, but less so.

Your assertions may be true for you, but they're certainly not true for me.

DriveThruRPG is a legal source of PDFs. A lot of systems have quick-start rules for free (for a try-before-you-buy deal), and WEG's D6 core books are also free. Some stuff is discounted, others... you still have to pay full price.

[Airk]

From your perspective, these things may be true.

Personally, I have trouble justifying $20 for a single book, and I'm certainly not accustomed to paying full retail. I suppose getting PDFs with the PirateBay discount is an option, but that's not one I prefer.

Uhm, getting PDFs for $10 or less is super easy and perfectly legal, as has been pointed out. If you have problems paying $20 for a book, how the HELL are you playing D&D? :P Since there, your choice are "LOTS of $50 books" or "Ongoing $15 a month subscription".

Or, you know, play a retroclone as Rhynn suggested, but which seems completely outside the scope of your experiences.

Personally, getting players together on a different system is a chore. Homebrewing rules is also a chore, but less so.

Have you tried like, actually explaining the situation? ^_^

Your assertions may be true for you, but they're certainly not true for me.

Actually, as folks have pointed out, many of my assertions are also true for you. This hobby is cheap.

[CarpeGuitarrem]

But basically, if this annoys you, don't complain about it. Go play something else. There's an infinite number of games out there, and basically none of them use d20s for everything except for D&D, it's "d20" derivatives, and Pathfinder, which is basically a d20 derivative.

Even better, in your particular case, you have a perfectly clear argument for why you don't like D&D: You don't like games with swingy results. So find your own game and enjoy it.

It's worth noting that many of those games only came into being because people saw issues with D&D and its kind, and worked to fix them. So I think there is a validity to this topic and discussion coming up every so often. I understand that it's a bit frustrating to see this particular one show up (since it gets newly discovered every month, it seems), but hey--everything was new to everyone sometime.

[Waar]

I've come to dislike the 'swingy' nature of the game when it comes to d20 rolls for determining success or failure of any given thing.

A d20 isn't exactly swingy on its own, not when it's used in simple situations where both succes and faliure are resonable, trying to armwrestle someone that is almost 4 times as strong as you is not such a situation, I don't think that should be such a check. (swingy, now that is flipping a coin: 50% chance to win, 50% chance to lose )

opposing checks

And seeing that an ordinary check wasn't random enough you brougth out the 2d20-21+mod (and who thought that 2d20-21+mod was ever a good thing to roll in a sysem based on 1d20-10+mod, oh right WoTC ). Opposed check are realy bad in their current form, a better way to represent an opposed check would be for the player to roll 1d20+(PC modifiers) and compare to 10+(NPC/misc modifiers). Now if you are in a situation where you want the modifiers to be more influential than the random factor, just double the modifiers: roll 1d20+2*(PC modifiers) and compare to 10+2*(NPC/misc modifiers).

[Airk]

It's worth noting that many of those games only came into being because people saw issues with D&D and its kind, and worked to fix them. So I think there is a validity to this topic and discussion coming up every so often. I understand that it's a bit frustrating to see this particular one show up (since it gets newly discovered every month, it seems), but hey--everything was new to everyone sometime.

I'm actually a little perplexed. There seems to be some misconception that I'm annoyed about the existence of this thread. I'm not. I gave an analysis of just -why- this sort of thing (d20 dice mechanics) is annoying, and then I gave a recommendation for what you should do if you find it annoying.

It's only when people started getting all "Don't you tell people to play other games! That's like work!" that I started to get annoyed. :P

[Knaight]

A d20 isn't exactly swingy on its own, not when it's used in simple situations where both succes and faliure are resonable, trying to armwrestle someone that is almost 4 times as strong as you is not such a situation, I don't think that should be such a check. (swingy, now that is flipping a coin: 50% chance to win, 50% chance to lose )

What about trying to armwrestle someone significantly stronger than you? Say, you can lift about 100 pounds over your head, and they can lift 130, which would be 10 strength vs. 12. It's close enough that it could potentially go either way, but there's an obvious lack of likelyhood.

As is, the stronger person wins 52.5% of the time and the weaker 42.75% of the time, with the rest made up through ties. The d20 is swingy here, with the linearity of it and the size of the modifiers contributing to said swing.

I'd also point out that if success and failure aren't both reasonable, there is a case to be made that the system should model that to some extent. Take FATE - an average person is going to have a 0 in Might. The strongest is going to have an 8. The opposed roll is 4 Fudge dice for both, which have a plus, minus, and blank side. The average person cannot win straight up, but they can tie, the odds of a tie are 1/6561. If they bring in some sort of way that it isn't a straight up contest wins become possible, though they aren't getting anywhere near likely territory.

[Rhynn]

It's only when people started getting all "Don't you tell people to play other games! That's like work!" that I started to get annoyed. :P

How dare you suggest that some people might enjoy some game they haven't played yet more than the one they don't like now!