Fix Event Weights After MadSpin
For compressed EWKino production, we'd ideally decay SUSY particles with MadSpin to account for spin correlations and ultimately give us the correct mll distribution.
In some tests, I've used MadSpin to perform the decays but I run into the notorious issue where the post-MadSpin cross-sections are tiny, which can be a problem for Pythia. In ATLAS, we have a recipe for fixing the event weights in these cases. I think the relevant function in MadGraphUtils
is here:
https://gitlab.cern.ch/atlas/athena/-/blob/21.6/Generators/MadGraphControl/python/MadGraphUtils.py#L1231
Can we somehow incorporate this fix into the workflow when running MadSpin?
If we can get MadSpin to be more reliable in the workflow, it would also be nice to improve the way that the MadSpin card is specified.