Physics World has brought us some fresh news about the Tevatron's future:
I understand all kinds of reasons why it would be a good idea for the Tevatron to continue. American particle physicists don't want to watch a relative demise of their place in the discipline. They don't want to lose their skills and speed - and jobs. And the collider works very well, indeed.
But if I rate the relative advantage of an operational Tevatron and an operational LHC, I guess that the Tevatron will be seen to be an obsolete gadget as soon as in 2011. Slightly higher instant luminosities won't be enough to beat the LHC's huge energetic advantage. And the latter can speed up certain (or most) discoveries (and the falsification of invalid hypotheses) above 100 GeV by a factor of 100,000 even when the luminosities are the same: see e.g. Alves, Izaguirre, Wacker 2010.
When the LHC accumulates 0.05/fb of data - and it already has collected about 10% of this amount - if will surpass the Tevatron with its 10/fb in the search for SUSY: check Jon Butterworth in the Guardian.
Also, people seem to use an incorrect argument about the Tevatron's integrated luminosity. It's much higher than the LHC's integrated luminosity, of course. But this is not an achievement of the Tevatron's operations between 2010 and 2014: it is an achievement of the Tevatron's work until 2010. ;-)
In principle, most of the acquired Tevatron data can be combined with the LHC data, too. The relevant quantity that should be compared if you decide which colliders should be running is the instant luminosity and energy - not the integrated luminosity. With this comparison, I am convinced that the Tevatron's operations will be seen as redundant ones in less than a year.