Micdiddy wrote:@melmoth: we have no need to assume that other things can avert the strike in order to call it flawed. The fact that the stimulus did not rule out this possibility is enough to call it flawed, without stretching so far as to assume anything. It's a mistaken negation, period.
Also, to address another point, I would be extremely surprised if the LSAC comes back and says "a flawed argument can be paralleled with a good one." I simply don't believe they would ever venture to make such a statement. This answer is NOT more right than any other answer. The contrapositive and the mistaken negation, even with quantifiers thrown in, cannot be misconstrued to parallel each other anymore than a raven and a writing desk. Either the LSAC will explain where we were mistaken, or admit they did something wrong (or just not respond).
I understand your concern and I see it differently. Let me propose a relevantly similar argument.
If I let the eggs rot then the room will smell but only if I do not throw them away. But based on the fact that it is (i.e. given, if, suppose) unlikely that I will not throw them away, it will probably not smell. So my point is the probability is calculated for when it is unlikely the eggs will not be thrown away. Of course you can say well it doesn't exclude the possibility that my brother passed gas and would make it bunk. But to that, I think LSAC could respond that is out of scope of the conditional conclusion, which applies based on (given, when, suppose, if) it is unlikely I will not throw them away. In other words, the argument is not assumed to apply when your brother passes gas--that is a different probability altogether. It is tangential to the initial condition that it is unlikely I will not throw it away.
You have to account for the fact that this stimulus is (1) conditional and (2) probabilistic. Given this, we can derive a conclusion when the probably sufficient condition is satisfied, but we can't conclude anything if we add alternate factors because that would satisfy another conditional and it would create another probability. You have to see that probability is in some ways a looser standard than formal logic and in some ways it is more precise and because it is not a guarantee it does not apply universally in every case.