Judging by what people stated in this forum, however, I just got the feeling that people will pick Tier 1 characters over Tier 2 characters almost all the time. Why would anyone settle for a less than perfect character if they had the choice to choose any character they wanted? And that's where I draw the line of what's fun and what's not. My perception is that some characters become more powerful because of a specific situation in the game while others become less powerful in the same situation. And some characters will work fairly well together and become more powerful in the process. And I'm getting the feeling that people don't appreciate or realize this.
Ronsen, I agree with your observation that our playgroups probably approach the game differently, which has something to do with why I wasn't able to offer much in response to your initial post, which I did find interesting. On the one hand, you offered all the OP asked for. Yours is an example of a group that has not left any characters behind as the power curve has shifted right with the expansions. Character selection, in your group, is a time of fun and mystery. The new characters, even with their obvious winners and losers, broaden your horizons, rather than narrow them. I'm glad to hear that and somewhat intrigued to learn about your group's approach to the game, which is clearly different from mine.
On the other hand, I harbor real reservations. When it comes to the fleeting excitement of flipping over that card and not knowing who'll be staring back at you from the other side, what's it really worth? For my money, the tingly feeling of walking into a blind date is a poor trade for the very real possibility you'll wind up stuck spending an evening with someone whose company you won't enjoy. Like Sam Anders, for example, who is a very selfish lover.
In all seriousness though, I think your idea of semi-randomizing the starting lineup is interesting and maybe worth a shot. I know a lot of people on these boards like Arkham Horror, which uses random character selection, and your modification would make this more like that. So I'm okay with the idea, even while I concur with Skowza that I enjoy choosing the team.
And then there was your big question: Are we reducing the fun and variety in the game by taking considerations of relative character value into account? My answer would be a categorical "no." I believe that considering the relative strength of a character when making initial selections is rational and responsible. Doing so reflects a recognition of, and respect for, reality, not a dearth of creativity. The cumulative effect of all this rational behavior is that the starting lineups begin to reflect a homogeneous quality, but I don't think the solution to that is for players to stop behaving rationally.
By way of analogy, if I could pick my starting hand in Texas Hold 'Em, I would very frequently pick A-A because it is the strongest possible hand. I would not always pick it, because there are a variety of factors in poker, as in BSG, that reward dynamic behavior over static behavior. Other hands I would frequently choose would be A-K, K-K, A-Q and occasionally Q-Q or J-J. I might even mix in a low-to-mid pocket pair or a suited runner-runner just to keep opponent's guessing and have a distant shot at stealing a surprise pot. The important point I want to make here though is that, if I could choose my hand in Texas Hold 'Em, I would very rarely, maybe even never, 2-7 off suit. I say this not because I lack the imagination to consider that the flop might come 2-2-7, but because I'm rational enough to take into account all the other possibilities that the table might present for which I'll be unprepared as a result of my poor hand selection.
One could look at the above analogy as an expression of a too-narrow thinking process. I certainly can't argue that in the strictest sense of the word, applying calculations of expected value to decisions is "formulaic." But I can tell you that I could play Hold 'Em for hours, days, weeks at a time, just as I could for BSG.
To be totally honest, I see my approach as an evolution from yours. At some point I outgrew looking for variety in the fleeting, superficial difference between, for example, looking at Saul Tigh's picture all night vs. Helena Cain's. Once I was liberated from looking for trivial distinctions between this game and the last, I began enjoying, *and creating* meaningful ones. There's plenty of variety in all the explosive and unpredictable events that arise from a dynamic board full of real and imagined threats, with an unknown number of hidden enemies seeking to deceive and destroy me. I don't feel I add much to the equation by trying to create superficial variety through making suboptimal choices for no other reason than I chose optimally last time.