People with a poor grasp of load fundamentals and anchor assessment can be expected to mis-apply things they read about on the Internet. Unfortunate but all too true.

Who stuck the defibrillator on this thread?

For the forces involved in top-roping, I don't see why a good old-fashioned tied cordelette isn't just fine. Much of the lengthy debate about anchoring methods was driven by the unlikely but not unheard-of catastrophe of total anchor failure when the leader falls directly onto the belay. There isn't much need to transfer those ultimate exigencies to mellow top-roping loads.

Nonetheless, just for the record, if you have three pieces in a horizontal crack linked with a tied cordelette and either of the outer arms fails, the load is not redistributed, it is applied entirely to the middle piece. The other outer arm will only be loaded if the middle piece also fails. For very large loads, this sets up the possibility of a cascade failure, in which each piece fails in turn at something near the total load, rather than their being and equalization of the load. These considerations really aren't (or shouldn't be) relevant to a top-rope anchor.

The sliding-X and other dynamically adjusting anchor set-ups are by no means the "only safe" way to rig. For the type of cordelette set-up described above, the tied cordelette typcially loads the center piece with half the total load, with the rest distributed to the outer arms. All of the sliding systems aimed at three-point equalization have so much friction that it seems possible that the three-point sliding systems are ultimately no better at equalization than the tied cordelettes. (Unfortunately, the resolution lies with drop tests that no one has done yet.) And as for the hybrid systems involving a two-point sliding-X or or improved variant like the equalette, they too load one of the pieces with half the total load and so are not clearly better at distribution than the tied cordelette, although they still retain the advantage or orientablility (again not of much concern for most top-ropes.)

har har har! it was not me!!!!!!!!!!!!!!!

Toprope anchors seem like a situation where the cordelette really shines. If the route doesn't wander, you don't need much in the way of dynamic equalization. And the monstrous redundancy of a cordelette is comforting when you consider that no one is up at the anchor to watch how it behaves when loaded.

At a belay anchor, you have a belayer standing there. If the anchor gets loaded in a funny way by the second or by the way the belayer hangs, and it starts rubbing against a sharp edge, you can do something about it. Any unpredicted problems like this at a toprope anchor could go unnoticed for too long.

I've been the proponent of a self equalizing version of the cordelette, but I use the old fashioned version more often when setting up TRs.

It was this time!