This is an interesting thread. For what it's worth (not much), Kweassa, I think you've gotten the better of the argument.
I've got relatively little experience with on-line flight sims: AH is my first and only, and I've been here a couple of years now. [Insert obvious disclaimer for lack of experience.]
It seems that the perk system is broken, because perk points and perk planes are relatively useless. The problems with perk planes, much discussed elsewhere, are the predictable result of the stigma that attaches to a plane once it is placed on the perked, non-free list (the perk-plane icons exacerbate these effects).
As I see it, the system will never really work until the perk stigma is removed (unless the goal is to essentially remove certain planes from the game, in which case they could simply be removed from the game). It seems to me that there are two ways to do this: perk nothing or perk (almost) everything. Kweassa's approach leans toward the latter, and I think that's the way I lean as well.
I realize this will never be adopted, but I think I would try a system something like the following: The perk prices for planes are adjusted periodically (e.g., daily) according to a heuristic that considers objective performance factors (e.g., level speed, climb rate, sustained turning rate, roll rate, range, WEP endurance, cockpit visibility, gun lethality, ordnance capability, etc.) and subjective performance factors (e.g., % usage and the various factors that comprise fighter and attack scores). I would not consider historical usage or date of introduction, as neither matters in the non-historical MA. The current country imbalance factor is layered on top, to really give that balancing mechanism some teeth.
What this would yield is a feedback system in which a plane's perk price is adjusted from a nominal, performance-derived value, depending on the relative extent and efficacy of its MA usage. The weight given to subjective performance factors is the feedback "gain" and determines how quickly and forcefully the system seeks to discourage (relatively) the use of common and/or effective planes. I would suggest a cap on deviation from the nominal values so that, for example, a Zero could never cost more than a P51-D, no matter how much or to what effect it is used. It might also make sense to place a deadband in-line with the subjective factors so that, for example, the nominal value of a plane would never be adjusted until its MA % usage rises above a certain level. (Actually, I would prefer a completely objective pricing criterion, but I realize that there are those in favor of diversity for diversity's sake. Also, there likely are some factors that make planes more or less effective than the "numbers" would indicate that would show up in the subjective valuation.)
I realize that there is the "what about guys who have no perks" problem, but that can be dealt with in any number of ways, such making all planes free for a month or two or three after subscription and then always maintaining a relatively small set of planes that are absolutely free. I would also make it easier to earn enough perk points to fly one's plane of choice. For instance, I'd select the nominal values such that just 1 or 2 kills of a "like" plane in a step-down model (e.g., La-5) would earn a flight in a step-up model (e.g., La-7).
If you don't like the notion of "forcing" people to fly something other than what they want to, then you'll just have to accept that there will always be a "Big 4." (Perhaps there will be no matter what you do.) Personally, I don't have a problem with that, and I'd suggest that as an alternative, just get rid of the jets and rockets and unperk everything else.
Just my $0.02,
JNOV
