One hotly debated topic for grognards of old-school D&D is which weapon damage system is superior: variable or non-variable. Tom Moldvay wrote the game using both methods, so I suppose even he was unable to settle completely on one or the other.
In Basic D&D, the rules were written with the option where all weapons rolled a D6 regardless of size, reach, weight, etc (Holmes only included this method). The basic argument here was that 1) the weapon's wielder was the determining factor behind a weapons deadliness, not the weapon itself, and 2) it was just easier to remember. The latter is just old-school silliness. If you can't remember you roll a d8 instead of a d6 for damage, but you still manage to remember your dexterity is 14 and your strength is 11, then I suggest you retake kindergarten math...the counting on your fingers part.
Anyway, the first argument has some merit, which I want to get to in a moment. But before that I want to talk just a little bit about role-playing a weapon. First of all, there is nothing stopping a DM from deciding that a halberd or battle axe is going to have a really difficult time swinging in a 5'x5' narrow corridor regardless of damage. There is also nothing stopping said DM from deciding that you can't attack from 6 feet away with a dagger. Oh, and before I hear someone retort, "Basic doesn't simulate reach", my answer is, "Basic doesn't simulate getting drunk at the tavern either". Just because the rules aren't there doesn't mean we don't and can't still simulate them all on our own.
A big part of Classic D&D is letting the DM decide what goes and what doesn't. As I explained in an earlier post about non-combat encounter resolutions, there isn't a roll for every procedure in the game. If you're into going by the book, play 3.5 or 4E. At the same time, however, we shouldn't have to throw the baby out with the bath water by abandoning all realism just for the sake of trying to out-old-school everyone else.
The goal should always be: how can I make this situation both as realistic and as simple as possible. In other words, I want to make this Battle Axe feel like a heavy killing machine, and I want the rules to be easy to use. There is a reason why there are hundreds of thousands of different medieval weapons, and not just "GENERIC KILLING TOOL A" - some were better at doing different things. Cave men generally didn't throw pocket knives at mastadons because long spears did more damage, in general, for all warriors in the tribe, even the unskilled ones.
You start cutting a potato with a big sharp knife, no matter how skilled you are with a butter knife. No matter how good of a coin flipper you are, if the coin is weighted on the tails side, statistically you will get tails landing up more often then down. It's called physics. And that's why non-variable damage is more unrealistic while at the same time providing very little to no simplicity advantages in the process.
Now let's get to some gaming examples. First, let's set the stage:
Two 1st level fighters wield a battle axe and a dagger respectively. They duel in an open grassy field with no environmental factors, wear no armor whatsoever and their to hit targets are equal. They have exactly the same amount of training in their respective weapons.
Now even in the best of conditions, in the real world, there is no possible way we can create perfectly equal conditions. But, by doing the best we can, we can create probabilities. You see this all the time on Myth Busters. As much as those guys annoy the heck out of me, it's based in real science, and it does actually work.
Really, the only thing left here is physics. Plain and simple. Imagine the two fighters swinging wildly at each other. Hits would be scored with an equal probability since our sample is equal, and I would imagine that neither would have a great chance of targeting "vital" areas because they are equally trained in defending those places. Now you're telling me that a swipe with a dagger would have the exact same effect as a swipe with a battle axe?
In other words, when all other things *are equal*, there is absolutely no reason for physics to be the ONLY factor that matters. The axe is equally sharp as the dagger, but is far heavier. An averagely swung slice with a dagger is likely to produce a nasty cut across the hand, arms, legs, face or shoulder. An averagely swung slice with a BATTLE AXE is likely to produce missing appendages. The former may cause the individual to swoon in pain, but applying pressure and bandages will likely heal it within a few days. The latter would produce instant shock, paralysis and severe bleeding - in all likelihood, death within seconds or minutes.
Bringing this back to real world executions, the purpose of them is to deliver the quickest and least painful death possible (at least in a humane society). There's a reason why modern executioners use a firing range, and medieval executioners used large axes or halberds rather than daggers. Physics. The scientific method requires a control and experimental groups. It's simple to understand why variable weapons work when they are tested against a sterile, controlled environment first like this example. Now put the dagger wielding thief in shadows, or in a narrow cave tunnel and his advantage becomes obvious. The DM should necessarily setup advantages for the thief, but that doesn't change the physics of the weapons, if it does happen to connect to a victim.
Another problem I have is when big monsters come into play. Many non-variable weapon apologists will argue that a weapon on the equipment list is just as good at killing another human being as any other. True, but this isn't Dungeons & Humans is it? Is a battle axe is really going to have the same effect as a dagger, versus a dragon? A dagger relies on hitting vitals to be effective. A dragon's vitals are FAR harder to reach than a regular humanoid all things considered equal. Like the mastadon example above, taking out large beasts requires larger weapons. Simply put: they do more damage. I don't go deer hunting with my .22 because it would take around 12 rounds to put the beast on the ground. Bigger bullets and more powerful guns cause damage more quickly than smaller bullets and weaker guns. It's all in the physics.
To further promote variable damage, I'd argue that it is more fun. Being able to throw a unique die for your character makes your role in combat more *unique*. If my weapon looks different from yours, it only makes sense for it to roll differently than yours too. The non-variable damage person will argue that it's all about the role-play, not the mechanics, but then why not play with non-variable hit points, non-variable ability attributes, non-variable AC, and non-variable hit tables too? Because it is, indeed, a lot about mechanics too. Perhaps for these non-variable apologists this is less about simplicity and role-play and more about trying to appear more "old-school" than the next guy.
And seriously, is it really that hard to remember what your weapon rolls?