One Thing That AD&D Got Right

[Desdichado]

However, the trouble was in AD&D was that those stat-bonuses *did* exist. Wish spells, magic items (girdle of storm giant strength, gauntlets of dexterity) and the like, and all of them were inconsistently applied.

For aesthetic reasons alone, I prefer level-based stat increases to the magic and item based stat increases.

I don't have a problem with either. It's just a blip on the radar of much more fundamental and difficult problems with higher level D&D.

 

[ggroy]

The +level/2 adjustment to attacks, defenses, and skills in 4E, is probably the easiest way of pacing the game to play like "always fighting orcs". 3E/3.5E and earlier TSR editions did this somewhat less precisely, but nevertheless achieved something like it for the first dozen levels or so.

The "always fighting orcs" trope essentially preserves the "sweet spot" of a 25% to 50% probability of succeeding in to-hit attacks, skill checks, etc ... when player characters are using their abilities linked to their primary stat(s). A success probability of less than 25% makes things somewhat frustrating, while a success probability significantly greater than 50% makes things somewhat easy (or trivial).

For example in 2E AD&D, the to-hit "DC" number to roll a d20 greater than or equal to for a fighter to hit a target (with AC_target) is:

21 - (level_fighter + AC_target)

The "sweet spot" of a 25% to 50% probability to-hit, corresponds to a "DC" number in the range of 11 to 16.

(For a fighter, THAC0 = 21 - level_fighter).

If one went through the 2E Monstrous Manual and figured out the fighter to-hit probabilities for various monsters, the probabilities which ended up in the "sweet spot" of 25% to 50% frequently corresponded to monsters whose number of hit dice were approximately equal to the level of the fighter. (This sweet spot would correspond to "level_fighter + AC_target" being in the range of 5 to 10).

 

[Dausuul]

What if the maximum normal benefit gained from a stat was a +1 or +2 at the extreme? What if the largest boost to actual ability came from gaining levels and devoting resources towards the skills the character wants to excel at?

I have lately been designing my own "personal edition" of D&D, kind of a hybrid of 4E and BECMI, and one thing I decided quite early on was that stats should not provide any modifiers whatsoever in the core combat mechanics. If you're a fighter of level X, your attack bonus is +Y and you deal weapon damage plus Z, end of story. Your Strength score doesn't change that in any way.

Stats in this system are used exclusively for "specialty" mechanics, like stealth* and poison, and for open-ended stunt-type actions. The primary reason for this approach is to eliminate the concept of "builds," but there are a couple other reasons too, and removing the incentive for stat inflation is definitely one of them.

[size=-2]*And even the stealth application is limited; rogues have a special "combat stealth" ability called Hide in Shadows which is not stat-dependent.[/size]

[Edit: Sounds like Swords and Wizardry works much the same way. Perhaps I should check it out...]

 

[ggroy]

I have lately been designing my own "personal edition" of D&D, kind of a hybrid of 4E and BECMI, and one thing I decided quite early on was that stats should not provide any modifiers whatsoever in the core combat mechanics. If you're a fighter of level X, your attack bonus is +Y and you deal weapon damage plus Z, end of story. Your Strength score doesn't change that in any way.

Stats in this system are used exclusively for "specialty" mechanics, like stealth and poison, and for open-ended stunt-type actions. There are a couple of reasons for this, but removing the incentive for stat inflation is definitely one of them.

[Edit: Sounds like Swords and Wizardry works much the same way. Perhaps I should check it out...]

I thought about doing something like this a year ago or so.

Except instead of dealing with the constant +Y attack bonuses to the players, I thought about moving most of that stuff to the DM side of the screen. A lot of the stuff to add up on both the player and monster sides, essentially "cancels out" in the final calculation in the to-hit rolls. The easiest way to remove all this extraneous calculations in the to-hit rolls, was to add a (level_monster - level_players) modifier to the defenses and attacks of the monsters.

The actual entries in the monster manual would have all the level dependent stuff stripped out (ie. +level/2, etc ...), but would state a "recommended level" for which each particular monster should be played at. The difference between the monster "recommended level" (level_monster) from the level of the players (level_players), is what gets added to the monster's attacks and defenses.

 

[pawsplay]

Stat increases can be a good thing if the rest of the system is built to be consistent with it. Two important questions are what exactly do the stats represent and how they affect tests during a game.

If we focus on the specific example of D&D, I agree that increasing stats is not a good idea. There are several reasons for it:
1. The advancement is already represented by other things (BAB and skill ranks in 3e, half level bonuses to everything in 4e). Increasing stats make it redundant.

Not in some ways. A higher state means, ultimately, a higher possible bonus. Without stat increass, such disparities are maintained indefinitely. That may be a preferable result, but it's definitely significant.

2. Each class has a narrow focus, making only 2 or 3 stats useful 99% of time. This encourages increasing high stats and ignoring low ones, making the spread greater with all problems that come from it.

I can't speak with authority on 4e, although I recall that mechanics make it possible to have two or three dump stats. In my 3e/Pathfinder game, this philosophy will get you killed. Perhaps it's a playstyle issue, but virtually every game of mine involves something like perception checks, grappling, a variety of saving throw attaks, and so forth. You can probably get away with dumping your two least favorite abilitys cores, maybe.

3. While high stats are harder to get at character creation, the difficulty of increasing them in game does not depend on stat value. It leads to the same issue as point 2.

That is something of an issue in that it's true the price doesn't go up, leading to min-maxing. However, the optimal spread is usually to spread your stats, rather than paying a premium on very high abilities in the beginning. Also, the relative disparty, if it continues, should remain about the same within a given class. That is, Int 18 wizard and Int 14 wizard remain even if they both focus on Intelligence. So, in summary, it shouldn't lead to the same issues as #2 unless the players decide to focus on a very specific style of optimization (one that in I feel in 3e does not pay off very well).

4. Nearly nothing depends on the stats alone. Because of that, the stats by themselves have no meaning. It's impossible to tell what given value of a stat represents, because in-game events would contradict it. That's an argument for removing stats completely, not only for not increasing them.

AC bonus, damage rolls, ability checks. I really feel like a 12th level rogue ought to have a better AC and Dex-based ability checks.

 

[Dausuul]

I thought about doing something like this a year ago or so.

Except instead of dealing with the constant +Y attack bonuses to the players, I thought about moving most of that stuff to the DM side of the screen. A lot of the stuff to add up on both the player and monster sides, essentially "cancels out" in the final calculation in the to-hit rolls. The easiest way to remove all this extraneous calculations in the to-hit rolls, was to add a (level_monster - level_players) modifier to the defenses and attacks of the monsters.

The actual entries in the monster manual would have all the level dependent stuff stripped out (ie. +level/2, etc ...), but would state a "recommended level" for which each particular monster should be played at. The difference between the monster "recommended level" (level_monster) from the level of the players (level_players), is what gets added to the monster's attacks and defenses.

Interesting approach... removes a substantial burden from the players but puts a slightly heavier one on the DM. Might be worth it though. Of course it also means you can't have class-based differences in attack bonus... I'll have to think about that.

 

[ExploderWizard]

I thought about doing something like this a year ago or so.

Except instead of dealing with the constant +Y attack bonuses to the players, I thought about moving most of that stuff to the DM side of the screen. A lot of the stuff to add up on both the player and monster sides, essentially "cancels out" in the final calculation in the to-hit rolls. The easiest way to remove all this extraneous calculations in the to-hit rolls, was to add a (level_monster - level_players) modifier to the defenses and attacks of the monsters.

The actual entries in the monster manual would have all the level dependent stuff stripped out (ie. +level/2, etc ...), but would state a "recommended level" for which each particular monster should be played at. The difference between the monster "recommended level" (level_monster) from the level of the players (level_players), is what gets added to the monster's attacks and defenses.

Interesting idea. I am using a similar mechanism in my own homebrew for saving throws. Every effect has an effect level. A monster/character of that level gets a standard save. Higher levels would gain a bonus and lower levels get a penalty.

 

[ggroy]

Of course it also means you can't have class-based differences in attack bonus... I'll have to think about that.

In practice, I've found most of the players in my 4E game typically used their primary stat attacks during combat.

It's when the players are not using their primary stat attacks in combat, is when my proposed system needs to have more stuff grafted on to it. (I never got around to completing it). For example, a 4E wizard trying to use a sword to attack, would be using their "dump stat" of 8 or 10 STR. One easy way of accounting for this in my proposed system, would be to give a -4 penalty to attack for the wizard attacking with a sword. (ie. This is the difference between an 18 INT primary stat mod of +4, and a 10 STR dump stat mod of 0, for a 4E wizard).

 

[Tequila Sunrise]

Also, I disagree with the idea that boosting stats makes bad stats stand out more. I mean it might make it more nociable that you've got no bonus or even a penalty on one or two bad scores, but even if you don't increase stats ever, the bad scores are still there. Mechanically a -1 from a stat is still a -1, no matter how many plusses you have, and whether or not you get more plusses from leveling.

What you say is true. However, I’m not talking about this from a strictly psychological PoV. Let’s take a simple opposed check example; a wizard and a fighter arm wrestling. At first level the wizard has an average Strength while the fighter has a high Strength. The fighter will probably win, but the wizard could get lucky and win.

Let’s say the two arm wrestle again at high level. The wizard still has his average Strength, while the fighter now has a really high Strength. Again, the fighter will probably win, but the wizard has to get even luckier to win. Hence, the low score stands out mathematically more even though it might not have changed one bit.

If you look beyond D&D you'll find plenty of examples for better ways to treat ability boosts.

Can you give an example? I’m familiar with WW’s point buy stats, but that’s it.

Another approach would be to replace the +1 to two stats with +1 to all stats. At that point, all stats would scale at an even pace.

This is what I’ve arrived at for my 4e game.

 

[Orius]

One would think that with his superior Intelligence, the wizard would know better than to arm wrestle the fighter.