While we're on the topic, there are many ways to 'cheat'.
Now, I'm not assuming that Leica is doing this--I, for one am assuming that this is raw lens performance, but until Leica publishes further information there will be no way to know.
But one can improve MTF with post processing, as David pointed out.
Eric, it depends on exactly what you mean by "sharpness" but there are algorithms which (with some 'hints') can restore a limited amount of resolution. You can Google deblurring algorithms to get some idea of what they can do, and their limitations.
MTF's can also be 'improved' by showing the MTF of the design (the design's theoretical best performance assuming perfect materials and construction), rather than of an actual, physical lens, with all its real-world flaws. (Zeiss is known for measuring the MTF of real-world lenses).
And other techniques involve using monochromatic light (light of only one frequency) to eliminate the all frequency-based abberations from the MTF chart. (Schneider is known for publishing the spectral distribution of the light source (ideally approximately flat from 400-700nm)).
And I'm sure there are other ways too. So, until Leica publishes the full information on their lenses, it's really anyone's guess. But given their reputation, I'm willing to give them the benefit of the doubt for the time being.
-Brad
That a very interesting idea, but how can sharpness be corrected in post? I'm sure Leica will be doing CA, distortion, and vignetting corrections since they were doing some of that already with the M8 and DMR but I think sharpness (contrast) can't be fixed with DAC? (Or if so how is it done?) And that's mostly what you see in the MTF. But you raise a a good point - we haven't seen the curves for distortion. Distortion is one of the trade offs in lens design. Maybe they were able to design the new lenses with DAC corrections in mind and concentrate on just sharpness? That would be a good explanation on how they were able to achieve those MTFs!