@Loon: My money would be on the LA in a dogfight as well given equal pilots, the point

made is that there is more to the game than dogfighting.
Agree the 88 is a bit of a headscratcher. It does have some interesting things going for it though that might partially explain things. #1 It is capable not only of level bombing, but dive-bombing as well, which is very cool and can give you more options and accuracy in a pinch. #2 It is torpedo capable. It makes sense that you'd pay a slight premium on options a given unit type has. #3 (And a distant 3rd at that) Fixed forward armament. You can actually theoretically dogfight from the pilot's seat in this bomber. Granted, I'd rather a TBM to an 88 any day, but it is a possibility.
@Save: Similar to point above: comparing the LA7 to another of the worst dogfighters in the game shouldn't be the criteria for getting something perked.
A distinction can be drawn between fighters and true dogfighters. I think we can agree the LA7 is more in the dogfighter camp. The fact is that most fights do not occur in an empty sector. In a many v many, factors like sustained turn lose importance relative to snapshot firepower and firepower duration. Aircraft that can blow away the enemy in the brief moment they end up in front of the nose are better off than aircraft that generally need to "latch" in order to get kills. (I put the Yak-9U in this category, Spits to lesser degree) Latching gobbles up SA and usually slows you down. I wouldn't expect the 190A8's MA k/d to fall below the LA7's any time soon.
@Messiah: Gonzoville is still pertinent as support to the degree it is accurate. Climb, speed, acceleration values are easily verified. If a given value didn't still stack up in game, I wouldn't cite it, just trying to save ppl time. Turn verification is a little harder but doable. Note that the climb times I cited were from my own testing tables (I'm slowly building an excel doc with performance values). Your point on lethality is valid when it comes to fighter-sized targets. For my testing purposes, I use hitting power at convergence. Yes, the hitting power of wing-based armament will be somewhat over-represented, but it is more practical than building, and averaging and/or weighing lethality tables for every 50-100yds of distance for the amount of performance differentiation achieved.
BTW (I actually docked the Corsairs when I was running numbers for my ENY idea due to convergence -gave 60pts vs 65 for Hellcat, should have done same for N jug but forgot...)