Here are my rebuttals:
1) Optimization means to make the best. To be best means to exceed all others. There is only one best. Which means to truly be optimized there is actually only one build per class that reaches the pinnacle of said optimization. Unless there is some other factor at play, such as teamwork.
By a clinical defination you may be correct.
But there are more than one way to define "the best".
The best at accurately hitting? The best at single target damage? The best at multi target damage? The best at skill use? The best at movement? The best defense? The best damage mitigation?
Every class has dozens of things they can choose to be the 'best' at. There is no single build to rule them all.
And yes, what may be "better" for you may be worse for others. For example the 4e warlord is the best leader at granting melee basic attacks. in a party of all casters, he wil NOT be the best leader.
As i mentioned, the teamwork factor is high, and group dynamics change everything. This is why the CharOp forums exist, and this is why the first thing that gets asked in every post is "What level, and what does the rest of your party look like?"
Optimization in 4e is essentially: making a effective charater, that is above baseline math, and matches the setting/campaign/rule restrictions set by the dm/players.
2) Since the players are capable of making 'sub-par' or less than optimal characters, either a) the mathematicians weren't very good. or b) they were good and INTENTIONALLY allowed for variety in character creation.
Also, unless A is true, the DM should be able to alter the encounters within the rules as written to accommodate for the party, if he deems that as a benefit.
Again, this ties into the last point.
By giving you a variety of ways to be good, they gave you the ability to make versatile characters and not follow the same rigid build.
If a caster wants to specialize in cold, there are good ways to do that. If another wants to specialize in fire, again there are good ways. Making optimal choices to fit one character concept in no way belittles the other options.
However, (for example) a wizard's main stat is Int. Every spell he ever casts uses a formula that includes his int modifier to hit the opponent. A player that has intentionally given himself an 8 int, will miss 25% more often than one that started with an 18, and can eventually get to the point where he only hits same level enemies on a roll of nat 20.
The one that started with an 18 can spend one feat, and will hit at roughly the same rate throughout his career. The second charator is more optimized than the first one. It's as simple as that.
3) While I do not concede your point that every encounter does not provide with an alternate solution, let us assume it is a true statement, ignoring the subjective portion regarding its entertainment value.
If this statement is linked with your statement, "Moreso than any other edition, the focus on 4e is team tactics/mechanics, and working together. A team is still only as strong as it's weakest link; everyone should have some degree of competancy," and also assuming truth in the statement about the team being only as strong as the weakest link, then our disagreement is in who the weakest link is.
I don't think there is enough evidence here to conclusively say, but it seems that the weakest link are the so-called optimized players. If the goal of the optimized characters was to support the concept of 'team tactics/mechanics and working together' then their failure to protect their allies is evidence of a flaw in their optimization theory.
Unless the optimized players are intentionally allowing the non-optimized characters die, which would be evidence of bias.
Again, 4e is very mathemtaical and statisical, and scales in a linear fashion. It is completely possible to design your character so incredibly poorly that a same level enemy can hit you on a roll of a 2, every round. If you have done this to yourself, there is very little your friends can do over the course of a campaign. Sadly, eventually the inevitable will catch up to you.
But i''ll try to address the various other combinations:
If the unop characters are similar to my wizard with an 8 int example it is clearly their fault since he intentionally made his character a detriment to the party.
If everyone makes smart enough choices to ride along (or ahead) of the default math curve of the edition, then the blame could be directed in multiple ways.
In either case it could be the dm's fault since he is designing encounters to match the power level of the optimized pc's, but not the unoptimized ones. This further complicates the math of the edition and makes the un-op players even more statistically useless than they already are.
Overall, it is everyone's fault for not agreeing on an op-level and playstyle to meet everyone's abilities and expectations before the game started.
FWIW [4e designer] baseline assumption was that roughly 70% of your feats would be put towards combat effectiveness, parties would coordinate, and strikers would do 20/40/60 at-will damage+novas. If your party isn't doing that... well, you are below baseline, so yes, you need to optimize slightly to meet baseline.
-Alcestis
