Madweight: hadronic taus

Dear Madgraphers,

I've been trying to run Madweight over some simulated z->tautau->lephad events.
i.e. as input, I have lhco events with 1 lepton, 1 hadronic tau (id=3) and some missing pT.

If I start a Madweight run using
generate p p > z > ta+ ta-, ta+ > l+ vl vt~
then Madweight will happily select events from the lhco and produce weights.

My question is about how Madgraph is interpreting these events.
From the diagrams produced and the Madweight log output ("2 missing particles", which I take to be the two neutrinos on the leptonic side) I suspect that Madweight is identifying the "id=3" object with the undecayed tau.
My simulated events have some missing momentum produced on the hadronic side, i.e. tau -> neutrino + "visible hadronic tau".
When working with hadronic taus, does Madweight allow for this neutrino, or is it identifying the id=3 object with the undecayed tau?

I'd be very grateful if someone could explain how this is handled, and how (if not like this) to persuade Madweight to allow missing momentum from the hadronic tau decay.

Thanks,
Ben