Here are my comments on the first draft of the note, dated Feb 11th. I have read all the document, but only got to write the first part. I will send the rest tomorrow morning (your afternoon).

I find the analysis in a good shape and the documentation is clear and rather complete. There are some sections which need a bit of work in clarifying, but nothing really major, I think. Please, find my detailed comments (line by line) below.

- As commented by Daniel for the 4l analysis, please indicate the final state in the title of the note.

Section 4

- line 160: Is it really like this? The LArHole is taken into account in the simulation directly? I thought a special treatment of jets and electrons needed to be done because it wasn't simulated. Also, in the line above, does "Boards" need to be capitalized?

Yes, the LArHole is taken into account in the simulation

Section 7

- line 240: The systematic uncertainty (not plural). Is there a reference for the 0.2% uncertainty in the muon triggers? (similarly as you give one for the egamma ones

added

Section 8

- The PV definition should come before, maybe just when d0 and z0 are first introduced in line 247?

done

- line 247: d0 and z0 aren't defined, say they are the transverse and longitudinal impact parameters?

done

Section 9

- line 352: You don't require OS leptons? Why?

Oversight, it is added to the selection requirements list.

- Table 9: why don't you add a row before indicating the number of events passing the trigger matching? This way one gets a feeling of the two lepton efficiency cut. And, why don't you add the statistical uncertainties? Also, caption: missing space between 4700 and pb

done

Section 10:

This is the section that needs most working, specially 10.4

- lines 473-474: but then this does not take into account possible changes in the acceptance of the WW background but just the overall normalization, right? Why is this background treated differently?

We are conservative with this background systematic, which is in any case, quite small.

- Table 21: The WZ SF is of 30% for the e-channel, but somehow I cannot seem to correlate this with the figures. Instead, there's this excess in the mumue channel but still the SF for the mu-channel is of 2%… Can you comment? the ee SF is summed over eee & eem. The mm SF is summed over mmm & mme. Maybe the log plots make it difficult for you to correlate it with the figures.

- lines 492+ : it would be useful to have a table with the chi2 and p-values, since it is very hard to read them from the figures.

- lines 495-498: Don't understand the statement made here: how is it possible these events would pass the event selection if exactly two leptons are required? Or you don't reject events which have an extra OF lepton?

This is for the 3 lepton control region, the same cuts are applied as for the signal region, but 3 leptons are allowed. The Z candidate is formed by taking the opposite-sign same-flavor lepton combination closest to the Z PDG mass.

- Fig. 12: assuming the mis-modeling is entirely due to the lack of QCD prediction, does the agreement gets better at high pT where QCD should be relatively small?

Yes, this is true. There are more plots now illustrating this.

-- MassimilianoBellomo - 15-Feb-2012