MadGraph5_aMC@NLO

Why is the BR of W bosons to leptons, when pushed through Pythia (and Delphes), less than the PDG value?

Asked by TMartin

(Background, ie: what I am hoping to resolve eventually) I was running a process using the MSSM, and I ended up getting approximately half of the number of leptons as I was expecting based on a rough calculation. I ran X1+X1- production, and let Pythia decay the charginos. The result was about half the number of leptons I would expect from a ~10% branching ratio.

Here is the X1 decay table.
# PDG Width
DECAY 1000024 1.21772961E-11 # chargino1+ decays
# BR NDA ID1 ID2 ID3
     3.74990888E-01 3 1000022 2 -1 # BR(~chi_1+ -> ~chi_10 u db)
     3.74990888E-01 3 1000022 4 -3 # BR(~chi_1+ -> ~chi_10 c sb)
     1.25009112E-01 3 1000022 -11 12 # BR(~chi_1+ -> ~chi_10 e+ nu_e)
     1.25009112E-01 3 1000022 -13 14 # BR(~chi_1+ -> ~chi_10 mu+ nu_mu)
#

(Issue, ie: what I think is a first step to resolve/understand before resolving my main issue) So I ran a simple process of on-shell, S-channel W production (p p > W+, p p > W-) and ran the result through Pythia/Delphes, allowing Pythia to decay the W. Again, the result was approximately half of the number of leptons I would expect. I ran 10000 events, and I get 659 electrons and 613 muons, where as the branching ratio of the W boson to leptons is ~10%. I checked the pythia .lhe file as well as the delphes .lhe file, and they give me the same results, so I don't think it is a mistake in my delphes setup.

Thanks for any insight you can provide.

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

Hi Martin,

Not sure that this is a MadGraph issue at all since you didn't ask MadGraph to decay the W.
(You can do it this way:
define W = w+ w-
generate p p > W , W > all all)
I would recommend to generate this way since then you have the full spin correlation.

Concerning your problem. My first idea concerning a factor 2 is that you forget to count correctly the anti-particles.
indeed for
generate p p > w+
add process p p > w-
you indeed expect:
621 +- 25 positron and anti-muon
[10000 events * 0.58 (relative importance of W+ compare to W-) * 0.107 (BR of the W-)
and
449 +- 21 electron and muon

Could you check that indeed you are checking for both signs electron, since your number looks compatible with only one flavor.

Cheers,

Olivier

Revision history for this message
TMartin (tmartin-4) said : 		#2

I wasn't sure who to address this to, since it is part of the package that comes with MadGraph. I guess I can pass this question on to the Pythia people.

The category in the Delphes output that identifies the type of particle doesn't distinguish between particle/anti-particle. It is a different column in the LHCO output that identifies the particle/anti-particle nature of the output. So I wasn't discriminating between particle/anti-particle in my check. When I checked with the LHE file, I used an abs() of the final state indicator to account for the particle/anti-particle difference.

I will go back and re-run, though, with W+ and W- separate and re-check, though.

Thanks for responding.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#3

Hi,

Indeed the way you perform your counting looks correct, But please double check it since this is funny that you arrive exactly on the expected number if you miss one. I really doubt that Pythia can make a so big mistake. But indeed if your result are confirmed then indeed you should contact them.

Cheers,

Olivier

Revision history for this message
Jeff Asaf Dror (asafjeffdror) said : 		#4

Hi TMartin,

My understanding is that Pythia has some sort of lepton tagging efficiency of ~0.75 due to default cuts that kills many of the leptons in the event. I'm currently struggling with this issue myself. Did you find a solution to your problem?

Can you help with this problem?

Provide an answer of your own, or ask TMartin for more information if necessary.

To post a message you must log in.