Results (the one w/o label are p_t binned, with pt = (v_h1+v_h2)_T [GeV]

Chi2 cut of 0.2<chi2/ndf/5.

The results I want to release are the first set with a cone cut of 0.3

With a one cut of 0.3, eta <0

eta >0

pT binning, eta < 0

eta > 0

Using a lower cone cut of 0.2: (only showing eta >0 )

Invariant mass binning (note y-axis range): (this plot accidentally includes PID cuts, next one is w/o)

w/o pid cut:

pt:

Loose cone cut of 0.4:

pT:

Asymmetries in eta binning. Note that the x value is not correct. In fact I used bins etabin1< -.5  < etabin2 <0< etabin3 < 0.5 etabin4

But it shows that the trend in eta is also roughly linear as expected since the p_t trend is linear and both have the same relation to x_bj.

I used the cut pt_comb > 4 and m_inv > 0.5