Disentangling Jet Modification in Jet Simulations and in Z+Jet Data

We study the impact of selection biases on jet structure and substructure observables and separate these effects from effects caused by jet quenching. We use the angular separation $\Delta R$ of the hardest splitting in a jet as the primary example observable. We first conduct a simplified Monte Carlo study in which it is possible to identify the same jet after quenching in a heavy ion collision and as it would have been if it had formed in vacuum. We select a sample of jets by placing a cut on their quenched $p_T$ and, as is possible only in a Monte Carlo study, compare to the same jets unquenched, and see that the $\Delta R$ distribution seems to be unmodified. However, if we select a sample of jets formed in vacuum by placing a cut on their unquenched $p_T$ and compare to those same jets after quenching, we see a significant enhancement in the number of jets with large $\Delta R$, primarily due to the soft particles in the jet that originate from the wake in the droplet of quark-gluon plasma excited by the parton shower. We confirm that the jets contributing to this enhancement are those jets which lost the most energy, which were not included in the sample selected after quenching; jets selected after quenching are those which lose a small fraction of their energy. Next, we employ a method that is available to experimentalists: in a sample of jets with a recoiling $Z$ boson, we show that selecting jets based on the jet $p_T$ after quenching yields a $\Delta R$ distribution that appears unmodified while selecting a sample of jets produced in association with a $Z$ boson whose (unmodified) $p_T$ is above some cut yields a significant enhancement in the number of jets with large $\Delta R$. We again confirm that this is due to particles from the wake, and that the jets contributing to this enhancement are those which have lost a significant fraction of their energy.