Beside each index and trait figure in the Euro-Star Index is the reliability percentage figure. An important element to interpreting the Euro-Star index on any animal (cow or bull) is to understand what the reliability figures mean. Put simply, the reliability figure gives an indication as to how much data has been recorded for a particular trait. For example, if a bull has a reliability figure of 99% on calving difficulty, then that bull is fully proven on that trait and the calving difficulty figure is based solely on the data recorded on that bull’s progeny. In contrast, a bull with a reliability of only 30% on calving difficulty has few if any progeny records and the calving difficulty figure is based almost solely on its ancestry (sire and dam). The higher the reliability figure, the less likely it is that the trait figure will change significantly in the future.
How quickly do reliability figures increase?
This all depends on the amount of data recorded on a specific trait and also the heritability of that trait. A highly heritable trait such as carcass weight will require approx. 100 records for a bull to hit 90% reliability. A lowly heritable trait such as calving interval will require a few thousand records to hit 90% reliability. The number of herds which data is recorded in also has an effect. For example, an AI bull with 200 progeny records coming from 50 different herds will have higher reliabilities than a stock bull with the same number of records, but all coming from the one herd. This is because the AI bull’s progeny are being born into many different herds with different management systems. This removes any potential environmental bias on the performance of the AI bull’s progeny through the power of numbers. It is for this reason that reliability figures of >90% are generally only seen in the Euro-Star indexes of AI bulls.
Understanding reliabilities is critical to the proper application of Euro-Star Indexes when selecting potential AI or stock bulls for your herd. More emphasis should be put on certain trait reliabilities in certain circumstances. For example, if picking easy calving AI bulls for use on maiden heifers, ensure that they are proven easy calving bulls i.e. >90% reliable. Where a reliability figure is <20% it means that very little or no data exists for that particular trait and there is more of a risk of that trait figure fluctuating in the future as progeny data becomes available. Genomics will increase the reliability figures on bulls before they are ever used for breeding and this will significantly lessen the risk of index fluctuation in the future.