I'd be sure to switch to the D4 as the tested camera (the link above takes you to the D300 tests, which are not indicative of FF performance).
http://www.dxomark.com/index.php/Lenses/Compare-Camera-Lenses/Compare-lenses/%28lens1%29/223/%28brand%29/Nikkor/%28camera1%29/767/%28lens2%29/320/%28brand2%29/Sigma/%28camera2%29/767#div1anchorIf you check the MTF charts, you'll actually find that the 70-200 VR holds its own against the Sigma resolution-wise. The Nikon is better on the short end, and, ironically, in the corners on the short end in particular (even though as Erik notes the vr 1 was probably optimized for DX which is all that Nikon offered at the time), and is also better at 200mm wide open in the center of the frame, though it rapidly falls to the Sigma's level as you move away from the center. The Sigma is a bit better at the long end at certain apertures, though the Nikon is still pretty close at the center. Where the Nikon will fall short is vignetting, where its APS-C optimization rears its ugly head, however the Sigma
also shows similar vignetting on the long end wide open (another unexpected result). You really need to dig through the DxO information to discover all of the relative strengths and weaknesses of a given lens.
There are two prior versions of Sigma's 70-200 f 2.8 on DxO Mark as well, and you might want to compare those, too. I actually think that the oldest version (their first "DG" version) may be the best of the Sigma lot overall.