Africa has a N-S structure, combined of the Nile and the Great Rift Valley complex. These two structures are connected at equator. The Great Rift Valley is often compared with Mid-Ocean Ridges, in the sense that if it would develop further, an ocean with a mid-ocean ridge in the middle, would eventually be formed. If we simplify these structures by drawing approximate N-S lines on a map (the 1977 World Ocean Floor Map created by Bruce Heezen and Marie Tharp), a curious pattern does emerge.

The structure fits into the pattern, slightly offset to the west if compared to the Mid-Indian Ridge. All the ridges sway a little, and the intervals are not perfectly regular. On the other hand, the intervals are regular enough to deserve attention. The ridges are all N-S oriented, the starting points of the N-S oriented sections are all found at equator, and the N-S sections are very close to being parallel to each other, separated by 90°.
Let us consider the possibility that a scientist, somehow curious about the structure of our planet, would look at the fourth N-S structure of 90° intervals. The said scientist would then find a main subduction zone, exactly at that location, found east of Australia, also mainly oriented N-S. Would our imaginary scientist keep on with the investigation? My wild guess is “yes”.
But somehow, some earth scientists would not continue with the investigation. They would refrain from it, because the regular distribution of mid-ocean ridges is contrary to the “gut feeling” that they should be randomly distributed. The precondition for that intuition is our moving picture of gradually drifting plates, where the ridges should continually shift location along with the plates. Thereby, most people will consider the possibility that the ridges had a different distribution in the past, and the distribution will be different in the future. Why should we then take the current distribution of mid-ocean ridges into account, when contemplating the science of Earth’s inner structure?
If this distribution is a coincidence, and we happen to live during the time of that coincidence, we still should be able to explain why this distributional regularity does “not count”. We can make some sort of an analogy to explain that. A clock has three needles, and we all know that the distribution is random, showing all types of patterns during a 12 hours cycle. Occationally, for instance at 12 o’clock, all the needles are unified. This happens literally only for a second, and then the random distribution of the needles becomes apparent again. So now we can ask ourselves, whether the distribution of those sections of mid-ocean ridges is of the same nature as when it is “12 o’clock”?
The answer is “no”, it would be too much of a coincidence. The solution to this problem is most likely that within the Earth a regular system of mantle convection is constantly moving the tectonic plates above, but at the same time the regularity of flow is manifested in some way or another. In a billion years, everything will have changed, but the system underneath will reveal itself, just as now.
Hope that this does clarify these things to some degree, as sometimes scientific findings can be explained much better with only a few extra words.