I literally just got through explaining this to people a few days ago, and then someone found this and linked it just now in relation to such.

Unfortunately, this is what you get when people have a love of “beautiful math” but they don’t actually understand the math they’re using and what it actually represents. As stated, the bell curve disappears as soon as you boil it down to a binary “success/failure” mechanic.

You DO get a bell curve with D&D’s 3d6 for attributes, because every single output is valuable. If you just take D&D’s 1d20+X modifier and turn it into 3d6+X, then the bell curve is only a mirage, and the only thing you’ve accomplished is making it more difficult to understand your chances of attaining that binary success.

What’s more frustrating… is they KNOW this is the case, because they balance the game around the % chance of each outcome. It doesn’t matter if there’s a 1d100 or 10d10, a 57% chance is still a 57% chance, and they balance their games doing exactly that, but then insist the bell curve is magically somehow making one 57% chance more consistent than the other 57% chance. No, it isn’t. You know it isn’t, or you wouldn’t have made it a 57% chance in the first place. What are you even doing?

It’s frustrating to watch, because they could do really interesting things with a bell curve, but instead all they do is create another linear 2-point graph except it’s now very difficult to tell what a +2 bonus gives and the GM can’t figure out what the actual difficulty is any longer without either a cheat sheet or a calculator.