Simple experiment:
Shinano upgraded to latest model, is modified in a way, that ALL her weapons are actually 12cm AA Rockets. It stays 2 hexes from Wake Island. Its airgroups set to standard training 40%.
Wake Island have large airfield, and additional B-29 unit (together with its organic WILDCATs), set to Naval Attack, at 1000 feets, 30% search. WILDCATs set to Naval Attack at 100 feets.
Date is early 1945.
Results:
During 11 Naval attacks, through several days, NOT ONE plane is reported as damaged, even when only WILDCATs attacks. Shinano uses large amounts of ammunition.
So, I have modified 12cm AA Rockets to have 10 effect. Results:
With similar number of attacks, around 30% also ends with no damages to planes, but rest have, on average, 3 damaged planes.
Only in ONE instance, there are planes lost to FLAK reported, and its only after training ZEROs from Shinano get caught into battle.
Quick research shows, that 12cm AA Rockets have around 80% explosives of 10cm AA Gun (so should have 22+ effect), so it is again modified to have effect 25. Results:
Most of attacks ends with some damaged planes. Still not ONE is reported as lost to FLAK.
Curiosity - original 12cm Rocket mountings (with 84 tubes, and ONE point of ammo) did not shot even once. But they were damaged, and even one of sides completely destroyed, so surely they had some planes within shoting distance.
As a side note - it was pretty rare, that B29 attack did NOT end, with at least 3 bomb hits. Crews were completely green, with 35 exp, and similar statistics in NavB, and LowN.
That clearly shows, that EVERY bomb makes its own attack roll. It seems we have similar problem, as with ASW, where it is number of weapons, what is most important.
I see similar solution, as I proposed for ASW - to change payload of larger bombers, with only FEW devices representing ALL bombs, so if plane carries 10 500lb bombs, it should actually use 2 * "500lb GP bomb (x5)" Devices, or even 1 * "500lb GP bomb (x10)". That way, against ground/naval targets, there should be only ONE hit possibility.






