Phrases like 'sheer stupidity' are getting thrown around a lot in this forum. Is it not more likely that if it is always 55% (and I don't know whether it is or not any better than you do), that there is a jolly good reason for it?
In this instance, it doesn't matter what the mechanical reasoning is. I don't care if it's for "making the math work", game balance, class balance, gameflow, speeding up combat, or any other imaginable reason. There is no possible mechanical explanation that compensates for the rediculously abhorrent effects such a thing would have on the verisimilitude of D&D. A flat 55% success rate for all saving throws does not belong in D&D. Why?
One of the core conceits of D&D is: Level Matters. While playing D&D, players assume that Level plays a real, logical, and consistent part in determining the outcome of events. The higher your Level, the better you are at doing things. No player would expect a Level 1 Fighter to kill a Level 25 Fighter under normal circumstances, just as no player would expect that Level 1 Fighter to have the same ability to shake off a Sleep spell that the far more experienced and hardened Level 25 Fighter has. Instituting a flat 55% success rate on saving throws would undermine the difference between a Level 1 Fighter and a Level 25 Fighter when it comes to resisting Magic, thus violating the core principle that Level Matters.
A flat 55% success rate on saving throws completely ignores Level, thus violating the verisimilitude of the D&D game world. If a Level 1 Goblin peon with 6 Wisdom has the same chance to recover from a Sleep spell as Asmodeus, Lord of the Nine Hells, then the system you are using to adjudicate such a thing violates the spirit of D&D. In fact, I would go so far as to say that any system that yields such a result is
not D&D, pure and simple. We can bicker about "Whatever Wizards Rubberstamps as D&D == D&D" all day long, but part of the core gameplay that makes D&D what it is does not allow for such a saving throw system.
So that's my reasoning. A Coin-Toss-For-Saving-Throws system does not belong in D&D for the aforementioned reason, hence my earlier proposal to houserule it away if Wizards feels compelled to include such an idiotic thing in 4E. If it is so ingrained into the 4E ruleset that it cannot be easily houseruled, the rest of 4E will have to pretty unbelievably amazing to convince me to buy the new edition.
That said, I'm a huge 4E fanboy and I can't wait until it's released. I'm carelessly optimistic that Wizards would never include such a boneheaded mechanic in 4E, or, if they do, I'm assuming it will be easy to houserule.