This was already explained by danny in post #326
It's caused by interference coming from other strong frequencies from 23.5E or 28.2E
When I first witnessed the discrepancy I had with the Mauritanian transponder on 12563 V being peaked better with the LNB on the T90 rail slightly west of the Badr 4 signals, I thought that interference from the Astra 2 & 3 satellites might have been responsible.
However, taking three transponders on Badr 4 (that are AFAIK still broadcasting on Badr 4 at the time of this post) in the case of the following...
12130 V (Dubai Media Transponder)
12169 V (Qatar transponder)
12226 H (Bahrain transponder)
...the Astra 2 satellites at 28.2E have UK spot beam transmissions on the same three frequencies & polarities, and come in very strong at my location. Meanwhile there is a transmission on Astra 3B at 23.5E on 12130 V but none on or around 12169 V or 12226 H - the 12130 V transmission at 23.5E still strong enough to be a nuisance.
On the other hand, there are no regular transmissions on either 23.5E or 28.2E that should cause local interference issues where I am receiving the 12563 V Mauritanian transponder, and it indeed it comes in quite well - Flysat suggests that there is a low symbol rate data transponder on this frequency at 28.2E, but this doesn't seem to cause issues for me receiving Badr 8 on this frequency.
When peaking the LNB on the rail for the three transponders mentioned above on Badr 4, all three had a common peak spot slightly east to where the LNB was peaked on the Mauritanian transponder on 12563 V on Badr 8 - the Dubai transponder had the lowest signal-to-noise ratio, presumably because it's dealing with interference from both sides, while the Qatar & Bahrain transponders are only dealing with interference from one side (at 28.2E).
Now, if both Badr 4 & Badr 8 were to be co-located to <0.1 degree, I would expect that the three Badr 4 transponders I mentioned above would peak better slightly more west than the Mauritanian transponder given the lack of nearby interference to the latter - but the opposite is taking place! At least for me.
From my experience above, I can only conclude that at present the Badr 8 satellite is currently sited slightly west to all the other Arabsat satellites present at 26E, for what reason I'm not sure.
In the core reception area of the satellites this is unlikely to be noticed (unless some very large dishes are being used that give a very narrow aperture to show this up) but is likely to be more apparent in fringe areas where bigger dishes have to be used and where alignment is more critical - it is similar to issues some people have mention on this forum in the past trying to receive UK beam transmissions in part of continental Europe with large dishes whereby Astra 2F & 2G are located at 28.2E but Astra 2E is located at 28.5E - my understanding with that is that Astra 2E has a Middle Eastern beam that uses the same frequency bands as the Es'hail 1 satellite at 25.5E, and that both operators need to ensure that the two satellites are kept a minimum of 3 degrees apart to reduce interference fro each other. Of course, in the UK & Ireland this 0.3 degree gap between the Astra 2 satellites isn't noticeable with the small (65cm or less) dishes in use here for reception from 28E.