The difference is not vanishing with tighter cuts.
Plots of several variables at preselection level and post final cuts show a disagreement between eeqq (4f) and ZZ+Zee (Pythia), namely a possible excess of Zee.
Check at generator level: sample of existing Wphact eeqq Lund evts (4 fb^-1) and newly generated (same luminosity) Pythia ZZ and Zee
No DELSIM: check on generation variables at parton level, selecting events with a e+e- pair and a q qbar pair.
Channel |
Generated |
eeqq sel |
cut1 |
cut2 |
eeqq 4f |
3999 |
3999 |
1728 |
1127 |
ZZ |
5352 |
821 |
276 |
224 |
Zee |
27240 |
17953 |
1225 |
701 |
ZZ+Zee |
32592 |
18774 |
1501 |
925 |
Cut 1: to get consistent sets (electron en. > 1 GeV, q en.>3 GeV, M(ff)>12 GeV, 5<theta el<175 degrees)
Cut 2: avoid 0 mass problems (ferm. en.>10 GeV + cut 1)
Numbers consistent after cuts (some gamma gamma in eeqq)
The distributions are not consistent!
A further selection requiring theta electron between 15 and 165 degrees (gamma gamma rejection) leaves 338 events in Wphact sample, 200 in the ZZ and 365 in the Zee. After this cut the discrepancy is clear both in number of events and in the distributions.