Some people think QCD measurements are only for "tuning Monte Carlo" or some such. In fact, that is just a small part of what we do.

Consider the inclusive jet cross section:

1. Verify accuracy of parton-level NLO calculations, which start to fail at low-Pt in our Run I experience. We now think hadronization is an important effect, and maybe we need NNLO if we go as low as 20 GeV.
2. Search for compositeness, etc., which would result in an excess at high-Pt
3. Our rapidity-dependent cross section drives the gluon pdf now at medium to high x. Before the PDF was very speculative because it was extrapolated from extremely low x.

We make measurements in a regime that is qualitatively well-understood but quantitatively at the frontier.

The dijet mass analysis is complementary to the inclusive, but has a slight reliance on Monte Carlo that we don't have with the inclusive. The dijet mass also has a greater sensitivity to new phenomena. We pride ourselves on nearly-complete freedom from MC in the inclusive (only the showering correction, ~1-2%, comes from MC).