K wrote:If real life was the model, the 14 STR guy would get beaten all the time. In the game, the difference is a +2. When we roll d20s, it will take dozens of rolls before we even realize that one character has a slight advantage over the other (even then, it might take a hundred rolls to actually have statistical proof).
You're right, but that's a problem with the statistical implementation, not the base concept of Character Attributes. In 4E GURPS, the difference between a 14 STR character and an 18 STR character is
extreme, because of how the rolling system works.
K wrote:Then we get to the question of "does this add anything to the game?" Personally, I'm fine with saying "all Fighters have a +4 to Strength checks" and have a feat called Great Strength that let's you use oversized weapons and gives another +4 to Strength checks.
And that would probably work pretty well in most games you play. However, some people would probably want an additional level of detail above and beyond what you want. The point to take away here is that the
system should support scalable levels of customization. Your example is perfectly workable by editing point buy at character creation to instead use a stat array by class and adding a feat - it's always easier to simplify a mechanic than introduce one.
Also, it's OK to say "I don't like this element of a system" without saying "This element of a system is horrible and needs to be removed for all players forever". It seems to me that the second term ought to be reserved for things that
really should be removed completely from a system - usually things like complete redundancies or intrinsically unworkable subsystems.
echo