Issue with cuts on pTW and mWW implemented in cuts.f

Asked by Ramkrishna

Dear experts,

I am writing to bring to your attention an issue we encountered during the generation of a gridpack for the aTGC measurement with a pT cut on the hadronic W boson at 150 GeV and the mWW cuts. Our gridpack generation was split into three mWW bins with common pT cut on 150 GeV, namely 150 GeV to 600 GeV, 600 GeV to 800 GeV, and 800 GeV to infinity.

Upon reviewing the pT and mWW distributions generated from the LHE file, we noticed that a few events fall outside the specified boundaries. Below, you will find plots illustrating this issue [1-3]

For your reference, I have provided links to the following relevant files:

1. proc card:

2. madspin card:

3. cuts.f card:

In the cuts.f file, you will find:

- A patch for the cut values for the pT and mWW samples on lines 86-113.
- The definition of the invariant mass function on lines 929-974.
- The definition of the pT cut function on lines 843-859.

Also, we tried to check the definition, but we cannot find the issue with the implemented functions in cuts.f

If you would like to know any other information, please let me know.

Best regards,


[1] pTW plot:

[2] mWW plot (linear):

[3] mWW plot (log scale):

Question information

English Edit question
MadGraph5_aMC@NLO Edit question
No assignee Edit question
Solved by:
Olivier Mattelaer
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said :

How do you perform such plot?
Such variable will not be preserve by the parton-shower, so if your plot are after parton-shower, this is perfectly normal.



Revision history for this message
Ramkrishna (ramkrishna-sharma71) said (last edit ):

Dear Olivier,

Thank you very much for your quick response.

I made this plot by grabbing particles based on the PDG ID. I get the final state electron, neutrino and intermediate W-boson that will decay to hadrons. Then made the mWW. For pT W, I grabbed intermediate W-boson using PDG ID that decayed to hadron.

We did not switch on the parton shower.

One more piece of information, we generated these events at NLO and with one additional jet (so, it may have up to 2 additional jets).

generate p p > ell- vl~ w+ $$ t t~ H QED=3 [QCD] @0
add process p p > ell- vl~ w+ j $$ t t~ H QED=3 [QCD] @1

Do you think this could be the reason?

In case you want to see the LHE file, you can find this here:

or from lxplus: /afs/

with regards,

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


I think that the "issue" is that you are plotting such variable after madspin.
If you were doing such plot before madspin, you will not get such issue.

So the cut are applied on the "production" matrix-element and within the MC@NLO framework.
So my bet is that at that level all your event pass (or not) exactly all the cut that you setup.

Then you pass the events to madspin. One of the feature of Madspin is to handle correctly offshell effect.
So the mass of the W are not exactly onshell (as they are in the production event) and is transform (in first approximation) to a breit-wigner. This transformation affects obviously other variable/... which explains that your cut are not preserved after Madspin.

In a way this is similar to parton-shower effect which also impact many observable and do not preserve partonic cut.



Revision history for this message
Ramkrishna (ramkrishna-sharma71) said :

Thanks Olivier Mattelaer, that solved my question.