Well, you raise a most interesting issue,,,
Firstly, when using the chirp radar I use a 3 element 28MHz yagi for transmit and a 2 element yagi for 28MHz receive at the remote. On 6m, I use 6 elements for transmit and receive.
That is the short answer, now for the issue of different radaition angles....
On 6 m and 10m, there is a vast difference in the performance of the home station and the remote. This difference in performance mostly favours the remote but sometimes the home station outperforms the remote. The radiation angles for 28MHz are mostly OK, I think, but only because of the small antennas, as you have indicated. On 6m the radiation angel (angle) effect is very evident, and sometimes stations with antennas at 4m above ground out perform the remote. Another observation is that 11 elements on 6m seem to mostly work better than 6 elements, better than the increase in gain would indicate.
So, I am coming to the conclusion that there is no shuch thing as the optimum station that will always
work better especially on 6m and maybe on 10m. On 20M VK3MO has demonstarted that a big stack does indeed out perform anything else, all the time, but that is 20m.
On 2m and above is another story altogether, see my article about this, see "Facts and myths about tropospheric ducting" at http://www.df5ai.net/Material/articles3 ... lesDucting
you can down load the complete article.
So, The radiation angles are a critical issueand one worthy of a lot more study.