the effect of the number of males
I haven\'t read the article you referred to, but the logic behind the idea is sound.
Very seldom do I breed one male over a group of females, although for experimental matings and in special cases I will mate one male over a special female or more. My usual practice is to cull most breeders, after the breeding season, and only keep a few of the best. I hatch and raise a large number of offspring and keep culling as they grow, once again keeping the best to use as breeders the following year.
The disclaimer is that there are always exceptions, but normally I put the best cockerels over the best hens and the best cocks over the best pullets. I try to use a minimum of two males over any group of females. One reason is to give me the highest percentage of fertility. The other is for genetic diversity. I generally change the males in the breeding pens on Sundays and Wednesdays, leaving the better male in for the longer 4 day period of Wednesday thru Sunday.
If a person only has four birds to breed from I can think of four ways to set up matings. The first and least effective way is to use either one cock over three hens or three cocks over one hen. All the offspring will share at least one parent and make them very closely related.
Another idea is to mate two pairs separately. This will give you two strains that do not share parents, but all offspring in each stain will be full siblings. There are some instances when this mating is most useful.
The best mating, using only 4 birds, to achieve the most genetic diversity would be to use two cocks over two hens. Put the hens in separate pens and rotate the two cocks between pens weekly or more often. This will produce some offspring that share the same sire or dame and some that are full siblings, but also offspring that won\'t share either parent. In this situation you wouldn\'t breed these offspring together (unless you know what mating each came from), but use the sire over the pullets and cockerels over the dames the following year.