The purpose of fletching is to help the following:
- Correct for inconsistencies in release
- Help orient the shaft to track more straight during windy conditions
- Correct for wind planing caused by the broadheads being inherently less aerodynamic/stable than field points
- Help to stabilize the arrows from archer's paradox as quickly as possible.
Of these functions, for traditional bowhunters, stabilizing the arrows as quickly as possible is going to have the largest effect on arrow performance as it'll be directly related to shooting smaller groups and directly related to improving penetration on game animals. As a general rule, more is more when it comes to fletching. A larger surface area of fletching will help make all of the functions listed above happen more consistently and make it all happen quicker. Larger fletchings will also slow your arrows down quicker (more pronounced arc to learn to shoot through) and create more noise.
If and only if your arrows are tuned perfectly to your bow and your accuracy/release is clean and consistent enough to know the difference between arrows being out of tune vs having a poor release, would I recommend going with tiny fletching. In order to cut fletching sizes down, however, I would recommend looking into and testing things like the amount of time it takes to stabilize your setup from archers paradox with paper tuning/video review, how much fletching it takes shooting in 10-15mph crosswinds to stabilize broad heads vs field points, penetration tests into foam targets at close ranges, etc. I would allow fletching size to be just larger than what your TESTS dictate is the minimum size you can use from your rig and still get good arrow flight with a broadhead and under poor conditions. In testing to do so, you'll have just a bit more steering than needed effectively minimizing weight on the rear of the shaft while simultaneously minimizing noise from the fletching but NOT compromising the primary functions of feathers. Until you and your rig is ready to really sit down and test the nitty gritty of above performance parameters, I'd recommend going with a 3x5" or 4x4" parabolic (quieter than shield) as those are pretty standard build parameters because they are what consistently works well for most people.