I think that plot is evidence that we got the beam correction right to high accuracy out to high ell. The noise in TT is not much worse than the noise in EE&BB - it's absolutely tiny compared to the signal. If we got the noise debias wrong by factor 10 then this test against Planck would likely still pass. However the ref's concerns are misplaced - we may unfashionably stick to using auto spectra but noise de-bias misestimation is demonstrably not a major problem. Clem -- ********************************************************************** Clem Pryke - Associate Professor - Physics University of Minnesota Room 318 Physics and Nanotechnology Building 115 Union Street SE, Minneapolis MN 55455 Tel: 612-624-7578 Fax: 612-624-4578 email: pryke(a)physics.umn.edu **********************************************************************