Mechanics of D&D/d20

[Lord Vangarel]

I don't know if it's the older I get or just too much time to think about things but I can't stop looking at the mechanics of D&D/d20.

As I see it the single d20 roll while an easy concept to grasp has a couple of major holes in the resolution.

Firstly, at low levels the modifiers are not very important. This, as I see it is because the d20 roll can produce any result as it is a single die roll so if you have the strongest 1st-level commoner in the world (Strength 18 modifier +4) arm wrestling the weakest 1st-level commoner in the world (Strength 3 modifier -4) the difference is 8. On a d20 roll the weakest commoner could roll a 20 modified to 12 and the strongest man could roll a 3 or less and lose. In reality I couldn't see this ever happening but the system allows it.

Secondly, at mid to high levels the situation reverses where the d20 roll becomes unimportant (see Deities and Demigods not even requiring rolls and Epic Level Handbook addresses this problem as well by altering the core mechanics for advancement). This situation occurs because a character with say a +17 as his primary attack can hit most creatures in the book without needing to roll so combat becomes who deals damage fastest. As the bonuses increase the problems continue to the point where Gods don't bother rolling dice anymore!!!

So the question I keep asking myself is what is the alternative? Don't get me wrong, I love D&D, I've played for 20 years and don't want to change the rules as such but the resolution of the d20 mechanic just seems a bit wrong! This isn't just 3E of course, it has affected all of the editions, I now know why my campaigns tended to break down at mid-levels upwards.

 

[Tonguez]

I tried using 2d10 at one point inorder to address those issues and get a more 'bell-curve-like' distribution (its not actually a bell curve but it does tend more to the centre then a linear d20

It was fine but the simplicity of d20 makes it much more fun and since I have a general cut off at Lv 10 -15 (ie in my games Lv 10 is considered high) the mega cores never really become an issue.

 

[Frostmarrow]

When the weakest wrestles the strongest there is a problem, I agree. However, this occurrs so rarely that the DM easily can make up house rules the day it's actually an issue. I always assume that the strongest will win any armwrestling contest. That is, a Str 18-guy will always beat a Str 17-guy.

When you gain levels the game changes. At low levels all combat is hit or miss but once you reach higher levels it's more about what you do to your opponent (damage, disarm, etc). This is a good thing because the obvious alternative is to make AC increase parallell to BAB and thus combat would always feel the same regardless of power levels.

Gods always rolls 20 on any checks (though they do roll to confirm crits). It's part of being a god.

 

[Drawmack]

The grim -n- gritty battle system handles this issue very nicely. It makes the modifiers smaller and everything is an opposed roll instead of just going against your opponents AC. So if someone has a high defense bonus they are still hard to hit.

As far as the commoners arm wrestling problem that you mentioned there is an intrinsic 8 point spread that must be covered. There are 400 possible die rolls (20^2 little set theory) when two people are rolling d20s against each other. Only rolls where the stronger commoner rolls 12 or less can be tied or beaten by the weaker commoner. Also the weaker commoner has to roll at least a 9 in order to tie the stronger commoner's 1. So we have a situation where the set has reduced to (12 * 11) 132 combinations possible combinations with enough of a spread that the weaker commoner can win. Now take into account that half of those rolls acctually have the stronger commoner with the higher score and you have (132/2) 66 possibilities that leave weaker commoner winning. So the chances of the weaker commoner winning are (66 * .25) 16.5%. This means that the payoff should be in the neighborhood of 6 times the bet if the weaker commoner wins. There is always a chance that the weaker commoner could win. Maybe the stronger commoner is hung over, not taking the challenge seriously, drunk, etc. This seems pretty realistic to me.

 

[Psion]

This, as I see it is because the d20 roll can produce any result as it is a single die roll so if you have the strongest 1st-level commoner in the world (Strength 18 modifier +4) arm wrestling the weakest 1st-level commoner in the world (Strength 3 modifier -4) the difference is 8. On a d20 roll the weakest commoner could roll a 20 modified to 12 and the strongest man could roll a 3 or less and lose. In reality I couldn't see this ever happening but the system allows it.

No, you are allowing it by applying the system blindly; the system does not state "use a straight roll-off for arm-wrestling". You assume that you would roll off an arm wrestle using the normal opposed roll system. I wouldn't. The system is useful but it is also simple; you can't apply it without using your brain. Were it not so, we would be flipping through the book for every specific implementation. As it is, the d20-roll is used to resolve a rather uncertain situation.

There are two possibilities here

First, you can just use the take ten rules -- the person with the higher strength always wins; resolve ties with a roll-off, perhaps using Con instead.

Second, when you have big differences, you can also make that difference more marked by requiring multiple rolls. For example, here's a quick arm wresling system:

Each character rolls a strength check. If a character rolls 10 more than his opponent, he wins instantly. If a character wins 3 more checks than his opponent, he also wins. You could throw con rolls in here, too.

at low levels the modifiers are not very important.
(...)
at mid to high levels the situation reverses where the d20 roll becomes unimportant

This isn't really true. Take a look at what you are really seeing here. The d20 roll has precisely the same impact at low levels at high levels. If you have a +0 modifier, a DC that is 21 beyond your modifier is as much beyond your ability as it is for a DC that is 21 beyond the modifier of a high level character who has a +25 modifier. The difference here is what DCs are challenging for a particular character; A DC 21 is beyond the first character and a DC 46 is beyond the second character. It is up the the DM (with a little guidance) to set the DC of tasks.

What you are seeing here is a design principle of D&D: levels are more important than attributes. D&D has always been a game where a sufficiently skilled human could take on giants; this sort of design is implicit in such heroic fantasy.

As the bonuses increase the problems continue to the point where Gods don't bother rolling dice anymore!!!

I'm not sure what you are talking about. If you mean due to their sheer skill -- well duh! Of course a god of the forge isn't going to have a problem forging a sword.

If you are talking about the divine ability to roll 20s for all things -- no. That is a special ability of deities and a significant boon at that. It is not because "there's no point." There is still a point when it comes to epic tasks.

I now know why my campaigns tended to break down at mid-levels upwards.

I'm not seeing the logical support for this statement.

Your 3e campaigns? My 3e campaign is still steaming along nicely at 17th level -- it's just a matter of matching the challenge to the character.

Earlier editions, my campaigns broke down around 13th level -- but that was due to the scaling (or lack thereof) in earlier editions and has little to do with the d20.

 

[Lord Vangarel]

Drawmack, nice analysis. Makes me think that we need reasons to explain when something odd happens (like a incompetant commoner out shooting an expert bowman for example) like he got lucky, or the expert had an off day etc.

Psion, I was talking about earlier editions. 1E tended to stop around 7th-9th level, 2E used to get to 11th-14th level, we're on the second 3E campaign, the first got to level 15.

As regards requiring multiple rolls to resolve something like the arm wrestle, yes we can but we're then using averages again which kind of negates my original point. If we use 3 contests to determine the winner (best of 3 wins the arm wrestle) it places the advantage back with the more skilled/better character which is perhaps what we want. What I tried t get across was that there's nothing in the books that sets this down.

 

[Psion]

As regards requiring multiple rolls to resolve something like the arm wrestle, yes we can but we're then using averages again which kind of negates my original point. If we use 3 contests to determine the winner (best of 3 wins the arm wrestle) it places the advantage back with the more skilled/better character which is perhaps what we want. What I tried t get across was that there's nothing in the books that sets this down.

Of course not -- the book is concerned with things like hitting monsters, finding traps, forging swords, and convincing the guards that a severed head in your trunk is really a stage prop -- things that are typically more important to a fantasy setting. I.E. more uncertain things. Getting to do contests with less variation require you to use techniques like the ones I mentioned before.

 

[wighair]

I agree that the basic mechanic CAN be a bit wonky, but this is a game and the wonkyness can be what makes it fun (ask Merry - WitchKing Slayer) Its the unusual results that people talk about for years afterwards. As long as there is some semblance of logic in there I think that is good enough. If you want a more balanaced curve you could even go for 3d6. The bottom line is d20 is not really simulation at all.

Sometimes the chances of something unusual happening are small even if the stats are close. Imagine two young brothers, one a couple of years older than the other. strengths might be 7 and 5. only two difference. how often you reckon the young kid will win? i'd say never. Which brings me to my second point... If it ain't gonna happen, don't bother rolling.

Mel.

 

[nsruf]

This situation occurs because a character with say a +17 as his primary attack can hit most creatures in the book without needing to roll so combat becomes who deals damage fastest.

I disagree with most of your conclusions for similar reasons than those given by the other posters. But the fact that AC increases much more slowly than to-hit rolls is something that bothers me about d20. Frostmarrow said he liked it that way, but IMO combat between opponents of equal power *should* be the same at all levels. For skills, you can scale tasks to any desired level of difficulty, but almost the only way to bring up AC to match BAB at high levels is with creatures having a huge natural armor bonus. I'd prefer a HD-based defense bonus or something along that line.