search

ytatus94 / Higgsino

The NUHM2 Higgsino study
0 stars 0 forks source link

Joey's mail 20170530 #14

Open ytatus94 opened 7 years ago

ytatus94 commented 7 years ago

Hi Yu-Ting,

We currently merge four DSIDs together for the Higgsino signal samples (whereas we originally did not). I think everyone filled in the cutflow before this merging, so if you require DatasetNumber == 393408, I think you might get a bit closer.

Once things look reasonable, make sure to document your findings on HIGGSINO-19!

Joey

ytatus94 commented 7 years ago

Hi Yu-Ting,

Thanks a lot for the careful checks! It looks like what I did in the cutflow using the framework here was missing an absolute value on the Dphi, which killed off all of the events with Dphi(lead jet,MET) < -2.5. That is 45% of the events, which means my yields should be larger by almost a factor of 2.

My entries also used the older b-tagging WP, which might explain why you have different numbers for the MET cut as well.

So I think what you're doing is correct, because the cuts all seem reasonable. I'm going to launch our new production today, and then I'll update the cutflow spreadsheet.

Thanks! Joey