The convoyed moves always happen after the fleet moves. That's how you generally resolve convoyed army paradoxes. Also, a convoyed army is considered to be coming from the unit the fleet convoying it is in and therefore cannot cut a support directed at that fleet.

I think this equivalent to the "second order convoy paradox with two exclusive convoys" of the DATC test cases. This website uses the Szykman rule, which would mean that nothing moves in this situation.

See webdiplomacy.net/datc.php

I think due to the simultaneous nature, that is going to be resolved as a standoff. But get one of the offline adjudicators and five it a try.

Here is a bit more explanation:
http://web.inter.nl.net/users/L.B.Kruijswijk/#6.F.23
http://web.inter.nl.net/users/L.B.Kruijswijk/#4.A

@Draug: this is a complicated second-order paradox that is not resolved by your earlier advice.

If you ignore the army convoys and just look at the fleet movements first, you'll see that none of the fleets are dislodged. However, if you allow the convoyed armies to move, they each would cut the support hold that leads to the other convoying fleet being dislodged. It would be paradoxical if neither convoy failed or if they both failed. However, the two consistent alternative resolutions of having one convoy fail while the other succeeds, would be undesirable, since you would have to give one convoy priority over the other in this symmetric situation.

This website adopts the Szykman rule (see 4.A.2 under http://web.inter.nl.net/users/L.B.Kruijswijk/#4.A), which dictates that both convoys fail in this situation, and hence no fleets are dislodged, and basically nothing moves. Other paradox resolution rules result in the same thing for this situation or do not handle this case. Note that even the most recent edition of the published rulebook (2000) leaves this situation ambiguous.

OK, thanks.

Yebellz, is it right to say that the armies don't move in a paradox, unless there is a single consistent solution where some of them move? Similar to what draugnar said.

i mean, the version of the rules which says everything fails in the case of a paradox does still mean most things which can move will move...

@orathic

Re: "is it right to say that the armies don't move in a paradox, unless there is a single consistent solution where some of them move?"

No. If I've understood your statement correctly, then I believe your interpretation does not conform to any of the paradox rules (including the Szykman rule used by this site).
See this test case:
http://web.inter.nl.net/users/L.B.Kruijswijk/#6.F.22
Here you have two consistent solutions, one where the convoyed armies don't move due to the fleets being dislodged, and the other where the convoyed armies do move and prevent the fleets from being dislodged. If you use your interpretation of the rule, since there is only a single consistent solution where some of the armies move, that solution would be selected. Note that none of the paradox rules actually select this outcome for this scenario. Also note that the adjudication procedure that Draugnar described earlier would be able handle this scenario by prioritizing the fleet dislodgements and preventing the convoys from happening, which is the opposite outcome from your interpretation.

Re: " the version of the rules which says everything fails in the case of a paradox does still mean most things which can move will move..."

I'm not too sure what you are saying here, as this sentence doesn't make too much sense: "everything fails" seems to contradict "most things which can move will move" (assuming that you are not talking about the other units that are not involved in the paradox). The "All Hold" rule for handling paradoxes is quite bad, since it doesn't agree with any of the other rules for the most basic paradox:
http://web.inter.nl.net/users/L.B.Kruijswijk/#6.F.14

I think the Szykman rule is better summed up as simply:

If a paradox arises, then all of convoyed armies that are involved fail to move and have no effect on their intended destinations.

In this statement, a "paradox" means any situation where there is not a single self-consistent solution (i.e., either 0 or 2+ consistent solutions). A convoyed army is "involved" if the other core orders in the paradox depend on the outcome of the move for its resolution. The DATC page describes more about how to technically calculate the "involved" armies as those that are part of the "paradox core", however it's fairly intuitive to identify them without having to get into the technical details