Mike the accuracies increase as more data for a particular trait is reported. The ratios for each trait are what the EPD's are derived from. Ratios that fall outside the expected range still count towards the accuracy beacause they will influence the actual EPD number. The other thing about the accuracy of the EPD is that is a common misconception is that it does not account for consistency. Two bulls could have an identical BW EPD and accuracy and one could have a much wider range of BW which would make him much less predictable.
I don't know how much the formulas or the models used for EPD's have changed or even how much the formulas vary between breeds. It sounds like the different models used to account for heterosis are using different values and I am sure they will continue to adjust these as they collect more data.
I don't know how much the formulas or the models used for EPD's have changed or even how much the formulas vary between breeds. It sounds like the different models used to account for heterosis are using different values and I am sure they will continue to adjust these as they collect more data.