[b]Azrai sez:
I prefer random-only rolls.
IMO the creation of unique characters are more important than the game balance.
In a really balanced game group all fighters have the same strenght score, same consti and same dex. Imagine a fighter with strenght 18 and one with only 15. The 18 should be an extraordinary score which is very hard to gain. And of course the 18 is unbalanced compared to the 15. I am against all kind of point-b. methods, they create stereotypical ST 18 and IN 8 fighters.[/b]
So, let’s see - you don’t credit people with the ability to think of unique or interesting character concepts of their own? Do you roll to determine which class and what skills your character has as well?
I don’t think you give the 3e point buy system a lot of credit there (have you looked at it?). While the system has its flaws, it does scale ability scores upwards. In the default value distribution, attaining a value of 18 is quite expensive, and will be a significant sacrifice in other areas.
Under the default 3e rules, the fighter you postulate would have ability scores that go like: Str 18, Dex 12, Con 13, Int 8, Wis 8, Cha 8. That’s one crappy will save, just for starters. That character will have +3 total for his ability modifiers. Different ways of spreading out those points might be: Str 16, Dex 12, Con 15, Int 8, Wis 11, Cha 8 (total modifier +4, better saves overall) or even Str 14, Dex 12, Con 12, Int 12, Wis 12, Cha 10. Which is the better spread of ability scores really depends on the campaign as well - if your campaign is a primitive hack-n-slash campaign, charisma and intelligence are likely to be worth significantly less than in a heavily intrigue-based campaign.
Of course, you can also easily modify a point buy system with simple rules such as “you must buy at least two odd scores, and no more than one pair of identical ability scores” or similar ways of making the scores more “interesting” and unique.
That said, I don’t use point buy in my campaign, and I wasn’t talking about a “point buy” system for bloodlines. A “point buy” system for bloodlines is pretty much impossible to rig - there’s nothing to “point buy” against. I remember someone tried to do it, which was, in my view, a failure - basically, “point buying” a higher bloodline vs number of blood abilities - which doesn’t make sense at all, since it leaves people with low bloodlines with many abilities, and those with high bloodlines with few. There’s not really much point in arguing about that, since what you’d essentially end up doing would be bashing a strawman - and a stupid one, at that.
My suggested method for picking bloodline strength goes like this (and it is in part D of the bloodline system suggestions that was posted recently):
-Any player can choose for his character to be non-blooded or blooded, and if blooded, he can choose a tainted, minor, major or great bloodline.
-Each bloodline level corresponds to a scion level - i.e. a great scion is a 4th-level scion.
-The character generates his bloodline score at character creation by rolling dice according to the number of scion levels he’s going to take, using a less random method than the 2e one - i.e. roll 2d4+4, 2d4+12, 2d4+20, or 2d4+28 (and there’s a feat that allows you to get a stronger bloodline score than this).
This method is essentially the same as the template method in the current playtest version, with the exceptions that 1) it does away with the bloodline ability score in favor of a bloodline spread identical to 2e; 2) it replaces the “empty” level adjustment with actual scion levels to boost the power level of scions a bit; 3) it allows PCs to start the game out with a set bloodline score - unlike the template system, a PC gains his full bloodline up front. In other words, it’s more like the 2e bloodline score, but with a more 3e-ish balance.
This model, or a similar one, is something I’d argue for - not a point buy system. If you wanted to use my method with a greater degree of randomness, you could rule that players that want blooded PCs have to roll on the BR Rulebook table to determine their overall maximum scion level.
I’m no proponent that there is one true path - a cat catches mice, regardless of its color. My point is simply that the default system should match 3e philosophy as closely as possible, and not be an attempt at wedging a 2e rules system into a 3e game, or a system that uses 2e philosophy disguised with 3e mechanical terms.