I overlooked the D&D model in my previous post on reconsidering skills, probably because I was deliberately moving away from it. However, for completeness I probably should include it for comparison. Rather than modify the existing post, I am adding it as a new one here.
From my previous post,
Goals and Desirable Characteristics of a Skill System
The design of skills in Echelon is based on the following assumptions and expectations.
- High-level characters are better at pretty much everything than lower-level characters, given the same training. Experience counts for a lot.
- For any particular ability, trained characters of a certain level will be better at what they are trained in than an untrained character of the same level.
- Training can overcome experience, to a point. Lower-level trained characters can expect better results than higher-level untrained characters (either based on check result or trained capability).
- Experience can overcome training in limited ways. Higher-level untrained characters can expect success more often than lower-level trained characters, if the difference in level is large enough.
- Natural ability (high ability scores) can provide some benefit but should be possible to overcome with a combination of training and experience.
- At any given level, in straight checks, trained and untrained characters should remain on the random number generator.
Echelon is largely based around such assumptions.
Let’s see how the D&D 3.x model stacks up.
D&D 3.x Design
A character may have skill points allocated to buy ranks in various skills. If a character has less than one rank, the skill is a simple check using the associated ability score modifier and success may be constrained. The character may be incapable of achieving certain results (such as a succeeding on anything more than a common Knowledge check). Skills are class (purchased for one point per rank, to a maximum number of ranks equal to the characters level + 3) or cross-class (purchased for two points per rank, to a maximum number of ranks equal to one-half (the characters level + 3)).
I won’t go into what happens with multiclass characters or how Intelligence ties into the skill system.
If similar rules (or at least skill bonuses) were applied in Echelon we might end up with a table of bonuses as shown below.
This design requires a little more examination than the previous two methods.
The first goal may be partly met. A fully-trained character (maximized bonus) is better than a lesser-trained character. However, an untrained character never improves. I do not consider this goal to be achieved.
The second goal is clearly met. At low levels the trained character has a bonus four points higher than his untrained counterpart (and has access to ‘trained only’ results). At high levels the divergence is even greater.
The third goal is clearly met. Since untrained characters never improve, training very quickly outmatches the native talent of the untrained character. Experience doesn’t even come into play.
The fourth goal is clearly not met. Experience doesn’t mean anything without training (and skill points allocated to the skill). No matter how many nights you spend on watch, unless you pay (possibly cross-class cost) your Spot skill will not improve.
The fifth goal is met as far as training is concerned. Native ability can match a small amount of training. Training quickly matches native ability (an average but trained person — 4 ranks — is as good as a gifted but untrained person… and is in fact better because he can also get ‘trained only’ results). However, again experience doesn’t come into it, so only training will let someone less gifted become more capable than someone more gifted. This one is partially met.
The sixth goal… not even close. At high levels the most gifted (Dex 9) untrained character can just match the least gifted (Dex 1) fully-trained character… one time in four hundred (9+20 vs. 28+1). Any other combination, though, they’re off the RNG. In fact, the best the untrained-but-gifted character can do with regard to fully-trained-and-gifted is 29 to 37 — still eight points shy of a tie, and that again is one time in four hundred. On average they will generally be about 27 points apart. I’m going to declare this goal totally unmet.
Given the goals described above, D&D’s skill system is far from successful. Two goals clearly met, two goals partly met, and two goals clearly not met. This approach is, for my purposes, clearly worse than either of the alternatives I considered in my earlier article.