TTZ-ZtoBB

[Cvico]

We've been requested to make ttZ ZtoBB enriched samples, based on the ttZ-ZtoQQ samples that we already have. The PR includes the json file for ttZ-ZtoQQ reuse of gridpack, and also a filter for the fragment that selects only events that do not have d, c, u or s quarks in the LHE.

do you think this is a good approach?

[sihyunjeon]

hmmmm do you think it's worth doing this? cuz it's Z->BB which is then going to be just 1/5 and in the end they do need to consider q->b mistagged bjets.

[Cvico]

I believe the main purpose of this sample is good stats for training some kind of ML for analysis. I'm not sure if they might do some b-ghost tagging or something to train (and therefore they may not care much about mistagg right?), but the main argument is I guess keeping good stats for that. So in terms of physics I agree it's not really that needed, but I thought it would be easy to include.

Nevertheless, if you and @agrohsje think it's better outside the scope, I think we can perfectly manage it within TOP and that's it! I made the PR just to keep the conversation here

[sihyunjeon]

I think we have to discuss where to draw the line

Do we take care of all the spinoff samples from our nominal samples?
Or do we draw the line based on cross section?

Let's discuss in skype, it could get lengthy