So basically the assumption is that a sorcerer is abusing polymorph with every monster from every book, planar binding, simulacrum, etc.? Everyone freely combines Eberron and Faerun material? What does bard do to justify its placement? I'm trying to figure out how you make a tier list of classes if all the game-breaking tricks are in. I'm not even sure what tiers of classes mean if the rules allow Pun-Pun, the Cheater of Mystra, etc., because then it becomes about ranking different broken tricks rather than classes themselves. I guess I'd say, then, that any tier list is simply going to be relative to background assumptions about RAW, RAI, and what the DM will and won't allow, so you have to spell out your assumptions.
Anyway, digging back into the other threads that this list is supposedly based on, I think some of this analysis is just wrong. The biggest problem is there's no real discussion of how well a class performs against equal-EL encounters at various levels: you measure D&D classes against the monsters. (There are some very weird campaigns I've seen that involve more characters with class levels and the like than monsters, but I've rarely seen the DM who has time to prepare optimized class-leveled characters for every session. Most DMs use monsters in most encounters.) Depending on whether you do it Playing-the-Same-Game style and match characters solo against encounters of EL equal to their level and see how many they win, lose, or stalemate or assume they're part of a party and analyze what tactics they could use and how effective those tactics would be, you'll get somewhat different answers but generally not so different as to move a class across tiers. If game-breaking tricks are allowed, any class with access to one of them will win any encounter of any non-epic EL solo. On the whole, as I said, I'd assign things to mostly some of the same tiers, but some classes are far from where my analysis would put them.
Anyway, I understand what's going on now. Thanks!