The plot above shows the study I made about the fraction f the jet energy contained within a cone radius R compared to R=1 (this I assume to be very close to the scattered partons energy). These results were qualitaitively similar to those published by CDF. See for instance the plot below taken from (http://www-cdf.fnal.gov/physics/new/top/2004/jets/cdfpublic.html)

Although I used SISCone for this study, I did not see any significant differences in the results using anti-kt or the kt alogrithm.

Then I tried smething REALLY simplistic. Let me assert from the above plots that a cone of R=0.4 (what Matt was showing I think) collects in 90% of the partons jet energy. I then simulate the partonic pt spectrum - at RHIC  dn/dpt ~ exp(-0.3pt)  - and what I would "expect" to find in a cone of R=0.4, i.e. pT_Jet = 0.9 pT_parton. That is what is shown below along with the ratio of these two spectra. Of course this is a gross over simplificiation of the problem but its illustrates, I think, that the fact the PYTHIA particle level is ~factor two off of partonic/NLO calc is not too wrong. See slide 21 of Jan's presentation to the JetVCorr this week. (https://drupal.star.bnl.gov/STAR/system/files/2015-03-10_jetcorr.pdf).