I'll agree on "should adjust upwards more" -- I keep coming back to the thought that the one big problem with the FRR wall is that it absorbs demand without reacting to it - no matter how many FRR swaps are taken, the rate doesn't shift upwards by a single iota until the entire wall is gone and the order-book on the other side gets swept through.

Yes it's populated by people who have abdicated any responsibility for picking a rate, so in a sense it represents supply at "any rate, I don't care how low", but I think it would make for a better simulation of a functioning and efficient market if we pretended that those funds belonged to people who did care, and who would ask higher rates in response to increased demand.

If for no reason other than the fact that, if ever we reached a point where everyone involved in the swap market decided to use FRR, it would just descend gradually to zero and stay there forever (or just be formally undefined) and neither of those sound like sensible ways to manage millions of dollars.

On your other points I'm less convinced; you do realise it's not set by an average of the swap offers, but by the average of active swaps, yes?