Sacrosanct
Legend
The preferred method for generating ability scores has been a topic of discussion in my group lately. Since we started playing 5e, I've mostly opted to use point-buy and allowed players to opt out and roll for their ability scores. In my next campaign, beginning this week, I've finally made the decision to outright ban random generation. I have some players upset with that, but I'm not moving back, and the group mostly supports my decision.
Anyway, after talking with other DMs and players, I decided to write about my thoughts on the matter and make it public (here), because maybe other people are interested in joining the discussion with different points of view (with or against random generation). If it reads just like a big forum post, that's probably because English is not my native language.
Cheers!
All but your last point is a person problem, not a problem with the random die rolling method. So all of those can be pushed aside as illegitimate reasons. Re: #4, that's true regardless of the method, and has little or no significant impact to actual game play.
We had this thread several months ago and when you run the scenarios, there's little difference between a randomly rolled PC and a point buy/array on average. The vast majority of random PCs fall in line with the other methods. Only a very small % of rolled PCs will be an outlier with several really high stats (or low stats). So unless you allow your players to keep rerolling stats to counter the odds, you won't even really notice it. And again, that's a person problem, not a method problem.
Also, when rolling, you have a higher chance of rolling lower than array or PB as well. And as you mention in #4, with bounded accuracy, each modifier means more. So rolling a PC that ends up with a 6 and 8 for two of his or her stats is pretty significant, and makes up for their 17 and 16 they rolled as well.
Every time this discussion comes up, most of the people arguing why random gen is bad either ignore that you can also roll lower than normal, and/or rely on people not following the rules. Either way, not a fair assessment of random PC stat gen.
Also, I don't buy into the whole "I feel cheated if someone else has higher stats." For one, it's a team game, not a competition between players. Secondly, in 99.9% of the cases, you wouldn't even know the other player had a higher stat unless you were reverse engineering their die rolls to figure it out. And who does that? Someone with issues, that's who. Even with array, no PC ends up with the same stats because the player chooses how to allocate them, they are impacted by race selection, and may be further modified by ASI or feat choice. If someone rolled a higher stat than me, I don't feel cheated any more than I feel cheated when playing basketball and another player on my team is better than me.
Bottom line: if you don't like it, don't use it. But don't tell other people not to use it because it's bad/wrong, especially on flawed and/or biased reasoning.