If the hammer is all the way forward, pressing in the firing pin, the flag cannot move. But if the hammer is in the rest position, where it could be driven by the backpressure of the primer, the vibration of the revolver (not recoil as such) could cause the flag to move upward. Now I agree that this requires a lot of things to happen the wrong way at the same time, but in fact that is sometimes the case when a gun is fired, and some are right odd.
The later flags I have seen have a forward extension on the flag that won't allow the flag to move upward when the cam is on the bottom (unlocked) position. (If you look, you will see what I mean.)
So is there a possibility of the gun locking up? Yes, everything has to be exactly wrong at the same time, but IMHO, it could happen.
There's a couple of factors you're overlooking here most notably inertia and friction. When the gun is fired, its recoil will drive the frame up and back toward the shooter. Because the lock flag is only attached to the frame rather loosely, it will resist this change in direction and thus be forced forward and down as the gun recoils thus pushing it out of engagement.
Also, as the hammer falls when the trigger is pulled, the friction of the side of the hammer rubbing against the lock flag will push it forward and down out of engagement as well. About the only way I can see the lock flag being pushed up into engagement without being acted upon by the cam would be if it were so poorly fitted that the friction between it and the hammer dragged it up and back as the hammer was being cocked.
Also, do you happen to know about when the forward extension was added to the lock flag? Both of the ILS revolvers that I have ready access to are 2007-2008 vintage and neither have such an extension.