Wrights Co-Efficient of Inbreeding - COI

The COI is  the expected percentage of homozygosity in a mating - or the average chance that the two alleles of a gene pair are identical by descent due to being inherited from a common ancestor

These are abstract values which are calculated assuming that genes follow the Mendelian Laws with dominant and recessive alleles. This is simplistic in breeding performance horses where a series of complex polygenic genes must come together to produce ability and conformation, but it can be useful for breeders to estimate the genetic relationship between two animals.

For the calculated coefficient of a pedigree to be of any use at all it must be based on several generations, not 3 or even 5 generations, but nine or ten. 

A 5-generation pedigree chart will often show no duplicated ancestors at all, while 8 or 9 generations  will show a considerably higher inbreeding co-efficient.

What looks like a horse with an outcrossed pedigree at 5 generations will likely appear as a well linebred horse with a concentration of closely related ancestors at 8 generations.

Having no common ancestors at 5 generations is no guarantee that common ancestors will not occur in great numbers further back - and the  mating can still achieve a fairly high inbreeding co-efficient
 

Even going out to 6 generations will often bring to light duplicated ancestors beginning on the 3rd generation which you would not see at 5 generations and certainly not 3

For example -  Furst Heinrich shows no duplications at 5 generations and looks like an outcross mating 

Going out to 6 generations shows he is 6  x 6 to Der Lowe and 5 x 6 to Romulus i

It also shows full siblings at 6  x 5 - Remus 1 and Ronda - by Radetsky out of Fidelia   -  the presence of full siblings will increase the COI considerably
 

Even a  common ancestor in the 8th generation will up the measurable amount of inbreeding in a pedigree.

If there are 2 or more common ancestors within 5 generations, the inbreeding co-efficient will be rather high
 

As Byron Rodger says - knowledge of the degree of inbreeding only helps if you know which genes have been concentrated in the pedigree.
 

Also, the inbreeding co-efficient will just add up the percentages for how much influence an ancestor  will have  based on how many times that horse appears and where it appears -  and its assigned percentage according to that position  – it is simplistic in this way.

It does not take into account horses in a pedigree that are prepotent or heavily linebred individuals

Ken Mclean said "Many breeders still believe in the  mathematic co -efficient table which displays a rapid decline of the genetic impact from ancestors in each generation. 

The hybrid horse need not transmit anything like the tables suggest. Prepotency will prevail over all mathematics.  One specific ancestor can mess up the statistics. We are kidding ourselves  if we think we can breed top horses by any kind of figure system"

Not all ancestors are created equal.    A very superior prepotent horse duplicated many times in a pedigree will not follow these carefully worked out percentages – it will contribute far more genetically that its measured position indicates.
 

Also, a common ancestor duplicated several times will pass on genetic material - but not necessarily in the same way that it was expressed in that horse himself – due to recombination and genetic variation factors.

The COI relies on paper records rather than genetic inheritance - it falls short in predicting the unique genetic makeup of an individual horse - it measures the expected inheritance of genes but does not account for random segregation of alleles and which genes will actually be inherited

According to Wrights theory a full brother and sister mating would have an inbreeding co-efficient of 25%

A half brother sister mating would be 12.5 %

Full siblings have the same pedigree based COI but may  differ up to 100% in their actual inheritance of specific genes from their ancestors

Wrights theory is poor at predicting which matings will produce superior athletes as performance relies on specific complex genetic combinations rather than just low homozygosity

A stallion or mare can be inbred but the level of inbreeding between sire and dam can be low

For example in For Romance i and ii

Dam Gesina is an extremely inbred mare with an inbred dam herself

Gesina is 3 x 3 to Donnerhall sex balanced

And 3 x 2 to three quarter brother and sister Sandro Song and mare Gaby on the falling dam line - by Sandro out of Gepard mares

Grand dam Gesine is 3 x 4 to full brothers Pik Bube i and ii

For Romance is by Furst Romancier who does contribute one daughter of Donnerhall on the  4th generation but that is it

The only meaningful linebreeding in 8 generations between For Romance's parents is that one line of Donnerhall so his COI should be fairly low - even though he is out of an inbred mare

For  Romance has proven to be a prepotent stallion that can stamp his type and sport ability even though he is still young 

This ability has nothing to do with his low COI and everything to do with his inbred dam