Shortly before I retired from TC I proposed that instead of using AMAN for the calculation of EAT's and sequence numbers, which it wasn't very good at, why didn't we use it for Estimated Landing Times and sequence numbers, which were generally quite accurate.
The usual reply was along the lines of "the ICAO standard is EAT's so that is what we'LL issue".
I know that if I was approaching LAM, tight on fuel, I'd be more interested in what time I'd be planting the wheels on the ground than an EAT that has often proved, in the real world, to be completely inaccurate due to the differing track distances from the 4 LL stacks and the vectoring methods used.