1 July 2013

Published Sires
How accurate are their EBVs

Criteria to become a published sire are given in the trait leader article. In summary, a published sire is either French Pure or Purebred, has had a calf in the past two years, and has an EBV accuracy of at least 75% in one of the three growth traits.

A review of the corrections made to the EBVs of these relatively high accuracy sires thus should provide an indication of the credibility and reliability of their EBVs and rankings.

EBV and accuracy information on published Limousin sires has been collected since December 2009. In the period to June 2013, 158 of the current 311 published sires appeared in the December 2009 listing.

Although many corrections could at first sight be described as spectacular or catastrophic, depending on whether the EBV "improved" or "worsened", in fact most corrections were within the bounds of error expected for the EBV accuracies given in December 2009. For example, typical maximum corrections were within the range one to two "standard errors" (refer to the ABRI website for an explanation of a standard error), which is probably a reflection of the EBVs being considered to be "reasonably accurate" at the time of their first publication.

More telling was the impact the corrections had on the EBV rankings of the sires. If the sires had been ranked in the top 10% of the breed, and marketed as such in 2009, then their fall in ranking would have ranged from about 15% (to top 25% of the breed) to as much as 85% (or bottom 5% of the breed)!

Many sires in the table are well known, or were bred by prominent studs. The fact that the sires' EBVs underwent significant correction suggests that too much weight was given by Breedplan to the sires' and their progenies' initial field measurements. Breedplan gives more weight to initial field measurements of animals bred by larger studs.

For animals that are not published sires, changes to EBV ranking arising from corrections to EBVs would be expected to be generally larger. This is the reason why stud cattle breeders should be extremely cautious to not mislead potential clients with animal descriptions that rely on ranking.

Contact us if you would like to know the identity of the sires.

Trait Advantageous
corrections
  Detrimental
corrections
CE Direct (%) SIRE A +4.2  -6.3 SIRE B
CE Dtrs (%) SIRE C +5.1  -6.6 SIRE D
GL (days) SIRE E -4.6  +2.0 SIRE F
BWt (kg) SIRE G -0.9  +1.3 SIRE H
  SIRE DD -0.9     
200D Wt (kg) SIRE H +10  -7 SIRE EE
400D Wt (kg) SIRE H +17  -11 SIRE I
       -11 SIRE J
       -11 SIRE K
       -11 SIRE L
600D Wt (kg) SIRE C +26  -12 SIRE M
       -12 SIRE N
       -12 SIRE O
       -12 SIRE K
Mat Cow Wt (kg) SIRE L -24  +47 SIRE C
Milk (kg) SIRE P +7  -9 SIRE Q
  SIRE R +7     
  SIRE S +7     
SS (cm) SIRE T +1.0  -0.7 SIRE U
  SIRE N +1.0     
  SIRE V +1.0     
Carcase Wt (kg) SIRE C +11  -12 SIRE W
EMA (sq cm) SIRE C +4.8  -2.2 SIRE X
Rib Fat (mm) SIRE Y +0.8  -0.7 SIRE D
Rump Fat (mm) SIRE Y +1.0  -1.1 SIRE D
  SIRE W +1.0     
RBY (%) SIRE Z +1.6  -1.1 SIRE AA
IMF (%) SIRE X +0.4  -0.7 SIRE BB
  SIRE J +0.4     
Docility SIRE E +32  -24 SIRE CC

____________________

The above is a personal interpretation of information freely available on the world wide web, and believed to be accurate. Alternative interpretations, comments, and corrections, are welcomed.


Comments

18 December 2013 | James

The changes in EBV's as more information is added is what is expected, especially as information from more herds is included. This is due to the performance of the sires progeny in a wider range of environments and management systems. The genetic potential and diversity of different herds is also captured.

This highlights the importance of as many herds as possible being involved in breedplan. The EBV's are estimates based on all available quality information available at the time of calculation. It is designed to inform decision making. Using sires with suitable EBV's but low accuracy is riskier than well proven sires with highly accurate figures based on many progeny in many different herds. However this does not mean you should not use young bulls with low accuracy, rather be cautious eg. do not use on heifers if you are concerned about birth weight, how much are you prepared to pay for a straw of a less proven sire?

At the end of the day nature works on variation to see what survives best and that is what breeders should be doing selecting what works best and make decisions based on all of the best information available.


19 December 2013 | Frank

James makes good points. The important thing is to treat EBVs only as rough guides to, and not precise indicators of, genetic potential. As the above article illustrates, even the EBVs of published sires (they have recent progeny and at least one growth trait accuracy of at least 75%) can change significantly as more progeny data is collected.


Post a comment

Click here.

Comments are moderated and will appear as soon as they have been approved. When posting comments you agree to be bound by our Terms and Conditions.