Dalek
I believe you are getting mixed up between “analysis” and “simulation” - there were points at which certain data were stored (recovered later from the actual equipment): waypoint change, last steering calculation and power down being most relevant to near the Mull; interpolation between these points is analysis rather than simulation.
When it comes to altitude, I believe simulation was used to suggest what their elevation profile could have been and I agree that this was contentious – I favour the Boeing analysis of the available data (by Mitchel - not the simulation) which, within the bounds of accuracy of the systems, includes in the possibilities that they had not increased in altitude significantly at all on that final leg from waypoint change to impact.
When you wrote <<The report shows the TANS to be working to a very high spec approaching the Mull. Seven satellites in solution and a doppler within 330 m of the GPS position. Overall accuracy would have been around 10 m, maybe even better.>>, this reinforces the point I made on this forum recently that there is a widespread misunderstanding of the way the navigation data is presented to the pilots – I will not elaborate just now as I had prompted a discussion on this and am waiting for someone to kick it off (although realistically I will probably have to do so) – for now I will just reiterate that the way the Doppler and GPS system data is combined to drive the Horizontal Situation Indicators (HoSIs) makes the system effectively unreliable to any useful degree of accuracy when approaching a landmark as they were, especially after a sea crossing, and that Flt Lt Tapper had expressed misgivings about this system's accuracy and would surely not have trusted it to have got in as close as they did, approaching high ground, by the time he changed the waypoint in that system (demonstrating that it was not of any use to them at that point).
It is one thing for the non-handling pilot (acting as the navigator, if you like) to be able to read accurate GPS data from the CDU, work out with reference to his maps and with his and the load master's eyes out where they need to go and relay this to the handling pilot while he flies the a/c visual say along a valley low level;
but it is quite another thing for the handling pilot to be able to keep on track by reference to his HoSI when cruising at speed on a long leg with no clear visual landmark to aim for nor any ground feature beneath him (road, river, etc as they were over sea) for him to correct for drift – he needs his HoSI over sea with a fuzzy mass ahead of him.