Originally posted by Anthony:
</font><blockquote>quote:</font><hr />Originally posted by tbeard1999:
If you're gonna use 2d6 (8+ being success), for instance, there should be a very small range of likely modifiers, because a mere +3 assures that you will succeed a great deal of the time (83%). Similarly, a -2 will assure that you will fail most of the time (83%).
In reality, there are plenty of tasks where a novice has a very small chance of success, and an expert is nearly guaranteed success. There are also plenty of tasks where the novice has no chance, and the expert has a lot of trouble, or where a novice can do it easily and an expert has no meaningful chance of failure. You just have to accept that some characters will be much better at specific tasks than others. </font>[/QUOTE]Oh I agree. I don't object to characters being good at some things.
My problem is what happens if the range of modifiers exceeds what the system can handle.
The most noticeable consequence in CT is that it is easy for almost *any* character to hit nearly all the time, if he chooses automatic weapons. And it's not terribly difficult to hit almost all the time with civilian weapons either. The same is true of NPCs, by the way. The reason is that it is too easy to get statistically significant bonuses. And while a clever referee can impose penalties to mitigate this, I resent being forced to do so.
And as noted before, bell curves reward relatively modest net modifiers and make really high net modifiers of minimal extra value. On a 2d6 system like CT, there's just not much statistical difference between a +3 (83% chance of success) and a +6 (100% chance of success). However there *is* a significant difference between no bonus (41%) and a +2 (72%).
And in combat particularly, it's very easy to wind up with net modifiers of 4+. With Book 4 weaponry, it's nearly impossible to avoid.
So my critique boils down to 2 things:
1. A bell curve system is far more granular than it first appears, due to the fact that net modifiers beyond a very modest amount produce little additional real world benefit. In the case of CT, a +3 makes success nearly automatic (83%).
2. This means that applicable modifiers must be extremely limited both in quantity and in size, unless you happen to like nearly automatic success or nearly automatic failure being common in games. I don't, by the way.
IMHO, these flaws cast serious doubts on the purported usefulness of bell curve systems in RPGs.
For a very simple example, consider a 2d6 system with 8+ being a success vs a 1d12 system with 8+ being a success. In both systems, an unmodified roll will succeed 41% of the time. However, look at how positive modifiers affect the chance of success (first is success chance in 2d6 system, second is success chance in 1d12 system).
+0 - 41% vs 41%
+1 - 58% vs 50%
+2 - 72% vs 58%
+3 - 83% vs 67%
+4 - 91% vs 75%
+5 - 97% vs 83%
+6 - 100% vs 92%
+7 - 100% vs 100%
See the difference? It only takes a +2 to get to a high chance of success (72%) in the 2d6 system, while it takes a +4 to get there in the 1d12 system.
Here are the success chances with negative modifiers:
-0 - 41% vs 41%
-1 - 28% vs 33%
-2 - 17% vs 25%
-3 - 8% vs 17%
-4 - 3% vs 8%
In other words, the d12 can accomodate a somewhat greater range of modifiers than the 2d6 system can -- about 50% more. (Not advocating a d12 based system for CT; just using it as an illustration).
This fact is neither good nor bad, but it has profound implications for the game designer -- a 2d6 system cannot handle a lot of modifiers nor can it handle modifiers that are much more than +1. Unless the designer is willing for virtually automatically success or failure to characterize the game.
Some have said, "well, the referee can just create some negative modifiers to offset high bonuses".
Even stipulating that this is so, there are 2 problems with this idea.
The first is that a good referee can make up for a lot of flaws in a game design, but this shouldn't excuse louse game design. Why should the referee be *forced* to do so?
The second problem is statistical. If you're gonna apply negative modifiers, you'd better be very careful because a very modest negative net modifier will drop the success chance to very low levels. In the 2d6 system, a -2 net modifier reduces the chance of success from 41% to 16%. A -3 modifier drops it to 8%.
What this means is that the referee may find it impossible to challenge one character because to do so will make a moderately less capable character effectively useless.
The reason is that the SAME modifier can have profoundly difference effects on characters that are only a few points apart.
Example -- you have two players both with assault rifles. Both have Rifle-2, but A has a DEX of 7, while B has a DEX of 8. A gets a +2 bonus to hit, while B gets a +4 to hit. In normal combat, A hits most of the time (72%), B hits virtually all of the time (92%).
You tire of this, and decide to create some drama in combat. You decide that they'll face someone who's gonna be tough to kill (though certainly not invincible). So, you have the next combat occur in a driving hailstorm at dusk. You decide that you'd like to reduce B to less than 50% chance of hitting, so you apply a -4 modifier.
B now hits 41% of the time. But poor A now only hits 17%
See the problem? The same penalty can have grossly disproportionate effects on different players that are relatively close in ability.
With a modest gap in proficiency, A guy is essentially rendered ineffective, while B guy is still effective. Should a 2 point difference be *that* significant?
In a 2d6 system, it is.
And the numbers don't get a whole lot better in a 3d6 system. That's why GURPS has to have defense rolls -- it's a trivial task to get a skill of 12 (74%) or 12 (84%) and hit most of the time. Of course, the same is true of defense rolls, which explains the clumsy rule that the defender must make his defense roll by more than the attacker makes his attack roll.
It works, after a fashion. But a better solution would have been to design a better system. By the way, GURPS works well with a d20 instead of 3d6. All of a sudden, high skill levels actually matter.
