The book AMERICA'S HUNDRED THOUSAND, by Francis H. Dean has a huge amount of data, including comparative data, on the US WWII fighters. It is an awesome tome. At the end, it has lots of comparative data.
Speed: F4U-1D is 10-20 mph faster than F6F-5 (depending on alt)
Climb: F4U-1D climbs better than F6F-5 (1-1.5 minute faster to 10k)
Range, no drop tanks: F6F-5 300 miles longer than F4U-1D
Roll rate: F4U-1D better than F6F-5 (by a lot at higher speeds)
Turning performance: F6F-5 a lot better than F4U-1D (see below)
Dive acceleration: F4U-1D better than F6F-5
Limit dive speed: F6F-5 better than F4U-1D (only by 6 mph)
Turn radius (scaled units, as estimated by speed for 3g stall at gross wt.):
FM-2, 12.0
P-63A-9, 14.9
P-61B-1, 16.0
F6F-5, 16.5
P-51D-15, 21.5
P-38L, 24.6
P-47D-30, 24.7
F4U-1D, 25.4
Note, this is turn radius assuming each aircraft can sustain a 3 g turn, no better, no worse, and so does not properly include the effect of engine performance. Still, it's interesting that the F4U-1D comes out so poorly in this. The book estimates that this is because the F4U-1D had a "spoiler placed on the right wing of the Corsair to eliminate an unsymmetric stall problem" that is estimated to drop the overall lift coefficient by a lot ("in fact an NACA test report notes this was indeed the case").
In the end, I think it boils down to the fact that the F4U-1D is faster and rolls better than the F6F-5, was much worse in lower-speed turning, but in real-life combat, lower-speed turning was not as important as speed and roll rate.
(As an aside, many people are mistaken about the idea that bigger, heavier aircraft have poorer turning performance. This is not what determines low-speed turning performance. If a big, heavy plane has a large wing with good max. lift coefficient, it will turn well, like the P-61 Black Widow.)