First, let's clear away the underbrush.
Fairness and participation
Fairness is a widely discussed concept. For instance, there is a vast literature triggered by the publication, over 40 years, ago, of Rawls's A Theory of Justice, which defends a conception of "justice as fairness". I don't know if you're familiar with that literature or not; I know it very well.
For current purposes, it's enough to say that what counts as fairness in relation to a particular social situation or endeavour depends, in part, on the nature of that endeavour.
The endeavour we are talking about - or, at least, that I am talking about - is a D&D campaign that may last for several years. I don't regard it as fair that a significant determinant of a player's ability to have an impact during that campaign should turn on a single set of preliminary dice rolls. The fact that every player has an equal chance of being stuck with an unfair outcome doesn't cure the unfairness.
Even those who play the game as fully or mostly cooperative can have a good reason to want every player to be able to contribute equally. It's a leisure activity, played for fun. Spectating, cheering on from the sidelines, or playing the sidekick, is often not as much fun as being an equal participant.
Mechanical difference and statistics
Some rolled characters will have significantly better stats than some other rolled characters (eg compare a rolled 18, 17, 16, 16, 12, 11 to a rolled 16, 14, 12, 12, 9, 8 ). Some rolled characters will similarly have better stats than points buy/array characters.
I do not care for this in my game. Hence I do not prefer rolled stats.
If the selling point of rolled stats was that extra high stats were set off against extra low stats, then you could just change the point buy/array options. (Although, given that a bonus in a primary stat is generally stronger than a numerically equivalent penalty in a secondary stat, this is not a straightforward design issue.)
But I'm not talking about the trade-offs in high vs low stats. I'm talking about the spread of actual values that results from rolled stats, namely, that some players end up with noticeably better stats, while others end up with noticeably worse.
In other words, it's inherent in rolling that some players will end up with better stats than others.
(If by "rolling" you meant rolling on a table of point-buy arrays, it would be a different matter. But by "rolling" I'm pretty sure you mean 4d6 drop lowest, or something similar.)
Someone who rolls stats and gets the 18, 17, 16, 16, 12, 11 result does not have to choose breadth vs depth. S/he has both. Another way of looking at it is that s/he has got multiple free feat/ability score gains for free.
Which is why I don't.
******************
Now that the underbrush has been cleared,
The actual issue
I prefer a game in which every player has more-or-less the same chance to mechanically impact the fiction via his/her PC. PC build rules that give some players mechanically superior characters to other players are at odds with this preference. Hence I don't use them.
There is no jealousy. There is no sour grapes. There is a preference for an equal capacity of players to play the game by mechanically impacting the fiction via their PCs. That's all.
(Your other number comments are various sorts of reiterated denials of the premise that one PC will be mechanically stronger than another. I have already responded to them above.)
Fairness and participation
I discussed this upthread, pre-necro.
Fairness is a widely discussed concept. For instance, there is a vast literature triggered by the publication, over 40 years, ago, of Rawls's A Theory of Justice, which defends a conception of "justice as fairness". I don't know if you're familiar with that literature or not; I know it very well.
For current purposes, it's enough to say that what counts as fairness in relation to a particular social situation or endeavour depends, in part, on the nature of that endeavour.
The endeavour we are talking about - or, at least, that I am talking about - is a D&D campaign that may last for several years. I don't regard it as fair that a significant determinant of a player's ability to have an impact during that campaign should turn on a single set of preliminary dice rolls. The fact that every player has an equal chance of being stuck with an unfair outcome doesn't cure the unfairness.
Even if this was true - and without more information we can't tell (eg what if Bob was engaged in insider trading?) - what does it have in common with a D&D game? A D&D game is not an investment, nor a gamble. It's a leisure activity that takes place over time, with the player character as the player's principal vehicle for participation in the game. If it's unfair for different participants to have vehicles of differing quality, an equal chance of having a good or a bad vehicle won't cure the unfairness.
Not everyone regards PC vs PC competition as toxic. (For instance, it was a pretty integral part of play in the game's early days.) In my game, the PCs frequently have different preferences for how NPCs should be treated, for which goals should be pursued, for how to prioritise matters, etc. Some of these differences are resolved by non-mechanical negotiation among the players. Others are resolved via mechanical play - eg a player declares an action for his/her PC in relation to an NPC. In situations of the latter sort, I prefer that the players come to the game on a roughly even mechanical footing.
Even those who play the game as fully or mostly cooperative can have a good reason to want every player to be able to contribute equally. It's a leisure activity, played for fun. Spectating, cheering on from the sidelines, or playing the sidekick, is often not as much fun as being an equal participant.
Mechanical difference and statistics
You keep talking about expected values. I have made it plain that I mean actual values.
Some rolled characters will have significantly better stats than some other rolled characters (eg compare a rolled 18, 17, 16, 16, 12, 11 to a rolled 16, 14, 12, 12, 9, 8 ). Some rolled characters will similarly have better stats than points buy/array characters.
I do not care for this in my game. Hence I do not prefer rolled stats.
You seem to be ignoring that I'm not talking about expected values. I'm talking about actual values. (If you don't understand the difference between expected values and actual values then you seem to be confused about how statistics works. 4d6 choose lowest, with 6 repetitions, is going to produce plenty of actual value sets that deviate markedly from the expected values.)
If the selling point of rolled stats was that extra high stats were set off against extra low stats, then you could just change the point buy/array options. (Although, given that a bonus in a primary stat is generally stronger than a numerically equivalent penalty in a secondary stat, this is not a straightforward design issue.)
But I'm not talking about the trade-offs in high vs low stats. I'm talking about the spread of actual values that results from rolled stats, namely, that some players end up with noticeably better stats, while others end up with noticeably worse.
No one is disputing this. The issue is about one character being mechanically stronger than another.
These two claims are false. The two styles do not end up with similar results. Rolling does produce more variation, and that variation has different patterns. Points buy or array builds end up with PCs who have breadth, or depth, but not both. Rolling allows lucky players to have PCs who have both breadth and depth, while leaving some other players with little of either.
In other words, it's inherent in rolling that some players will end up with better stats than others.
(If by "rolling" you meant rolling on a table of point-buy arrays, it would be a different matter. But by "rolling" I'm pretty sure you mean 4d6 drop lowest, or something similar.)
Assuming that the array rules are properly designed, and assuming that your players' build choices express their preferences, this is unremarkable. One has chosen depth, the other breadth. It's like choosing a feat over a stat increase.
Someone who rolls stats and gets the 18, 17, 16, 16, 12, 11 result does not have to choose breadth vs depth. S/he has both. Another way of looking at it is that s/he has got multiple free feat/ability score gains for free.
But getting 18, 17, 16, 16, 12, 11 from rolling, vs 16, 14, 12, 12, 9, 8 is not a choice. It's luck. I don't see how extolling the virtues and consequences of choice in PC building gives me any reason to want die-rolled stats.
I'm not arguing that die-rolled is worse for you. Of course I'm arguing that it's worse for me - if it wasn't worse for me then I would have no reason not to use it! But it is worse for me, and hence I do have a reason not to use it.
Which is why I don't.
******************
Now that the underbrush has been cleared,
The actual issue
OK, I've bolded the relevant bit. This is what I have been saying. [MENTION=9053]SteveC[/MENTION] has said the same thing. It makes all the preceding attempts to argue that there is no mechanical impact of rolling largely irrelevant.
Correct. I don't understand why it "chaps your hide" so badly that some people do not want these sorts of variations in mechanical effectiveness among PCs as part of their game.
Let me reiterate the argument, then.
I prefer a game in which every player has more-or-less the same chance to mechanically impact the fiction via his/her PC. PC build rules that give some players mechanically superior characters to other players are at odds with this preference. Hence I don't use them.
There is no jealousy. There is no sour grapes. There is a preference for an equal capacity of players to play the game by mechanically impacting the fiction via their PCs. That's all.
Correct. This is my argument. It involves neither jealousy nor sour grapes.
(Your other number comments are various sorts of reiterated denials of the premise that one PC will be mechanically stronger than another. I have already responded to them above.)
