EPDs are predictions, and while an animal's true genetic merit will never change, it is safe to assume that their EPDs may change from one calf crop to the next, as more information (i.e., progeny data) is evaluated. A possible change value is an estimation of the change potential (in trait units) for a given EPD. The true genetic value of each genetic prediction would be expected to fall within (plus or minus) one standard deviation 2/3 of the time. (The other 1/3 of the time, bulls EPDs could change more than one standard deviation.) The possible change chart (below) shows the potential variation in EPD values that could be observed at different levels of accuracy for the various traits. For all traits, as accuracy increases the possible change value decreases and vice versa.
Assume a 60 lb yearling weight EPD had an accuracy of .30. We would expect the true genetic value to exist between 42 and 78 lbs two-thirds (2/3) of the time (60 lb YW EPD +/- possible change of 18). This is what is referred to as the confidence range. As accuracy increases, the confidence range narrows.
For example: if we increase accuracy in the above example to .80 acc., the true genetic value would be expected to fall between 55 and 65 lbs two-thirds of the time (60 +/- 5). Confidence range estimates the potential variation between an animal’s published EPD and their true genetic value. It does not imply more or less phenotypic variation in an animal’s progeny. This means that a low-accuracy sire should not be expected to have any more or less variation in his progeny than a high-accuracy sire.