How DV slots for regions are calculated
It does sound complicated
It is simpler than it seems. The following is the gist of it:
All the regions are divided into "high admission" regions and "low admission" regions based on how many immigrants they sent over the previous 5 years. This depends on whether the number of immigrants that came from a region is less than or greater than one sixth (1/6) of the total number of immigrants from all over the world. Under present trends, invariably Africa, Europe, North America and Oceania get classified as "low admission" regions while Asia and South America get classified as "high admission" regions.
The percentage of DV slots that are alloted to all the "low admsion" regions collectively is made equal to the percentage of immigrants that came collectively from the "high admission" regions over the previous 5 years. The rest of the DV immigrant slots are alloted to "high admission" regions.
Just to illustrate this point assume that over the previous 5 years the number of immigrants that came from Asia and South America is 70%. Then 70% of the DV slots are alloted to Africa, Europe, North America and Oceania. The other 30% is alloted to Asia and South America.
It's clear from this how the regions that sent lots of immigrants over the previous 5 years are disadvantaged in in the DV program. Their collective percentage of DV slots are are determined by
inverting the percentage of immigrants that they sent.
The visas that are alloted to "low admission" regions
are divided up among those regions in proportion to their populations. The same is the case for "high admission" regions.
This is just the basic idea. The page I linked to specifies the exact way the calculation is done. For example, those that came through the DV program itself or as asylum applicants are not included when making certain calculations. Also the population of countries that sent more than 50,000 immigrants are not considered in the calculations involving the populations of regions.