The way I'm seeing it (and I'm no flight expert), it's because satellites pick up signals in a circular fashion around them (see this map):
Pings ("handshakes") between the satellite and the plane establish that it is a certain distance traveling either northward or southward along an arc. Radar tracking (which is a different form of tracking) helps pinpoint which one of the arcs it's on (if radar is available).
The only radar we've had reported (IIRC) is the Malaysian military radar which last tracked this plane going NW towards the Andaman Islands. The Indonesian military radar did not pick up the plane. To my logic, it would seem, then, IF the plane traveling in a steadily receding ping route (longer pings) without change of direction as was established at the end of the last thread, that there is still cause to think the plane may have traveled on the northern arc route, as the plane's last documented direction (radar) was in a NW'ly direction. There has still been nothing to say it couldn't have, in other words, and certain info now to suggest it should have (ping data).
IMO (more in the theory thread), I believe the complete lack of radar data after passing through Malaysia military radar may be due to advanced radar-blocking abilities used by the hijackers.
(Personally, I think the 'debris' some satellite images are picking up down by Australia are just part of that part of the ocean's reputation for being the 'garbage basin' of the sea. I could be wrong, though, and my knowledge in this area is certainly limited.)
Nice theory and explanation... thank you, I agree on both the direction and the debris seen. Is the graphic on the left hand side from actual or hypothetical (to some degree) pings/radar?