Menu
Forums
New posts
Search forums
What's new
New posts
New media
New media comments
New profile posts
Latest activity
Media
New media
New comments
Search media
Members
Current visitors
New profile posts
Search profile posts
Log in
Register
What's new
Search
Search
Search titles and first posts only
Search titles only
By:
New posts
Search forums
Menu
Log in
Register
Forums
Cattle Boards
Breeding / Calving Issues
Are we too dependent on EPD's?
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Help Support CattleToday:
Message
<blockquote data-quote="simme" data-source="post: 1727203" data-attributes="member: 40418"><p>So where do EPD's come from and how are they calculated? Random number generator? I don't think so. It is an estimate, an expectation. It is not a promise or a guarantee. It is like a forecast based on data. </p><p>Take a trait that is easily measured like birth weight. You just need a scale. Simmental has a planned mating EPD estimating tool. Just plug in the sire and the dam that you plan to mate and hit the button. It is mostly just an average of the EPD's of the sire and dam. (I say mostly because there may be some "expected" adjustment based on heterosis depending on the mating.) We know that the actual mating will combine genetics from the parents with some variation. There is no equation that calculates the birth weight of the calf from the EPD's. But if both parent families have heavier than average birth weights, isn't it logical that there is a high probability that the calf they produce will have a higher birth weight than average for the breed? But also possible that it could be low birth weight? </p><p>EPD's will predict a higher expected bw. If the actual bw from the scale is lower than breed average, is the EPD wrong? Maybe or maybe not depending on environment. Let's ignore that environment thing for now. </p><p>So bw epd is high and the calf is one of the lightest in the pasture. Well maybe that EPD <strong>is </strong>wrong. Let's look at the accuracy. Say this is the first calf from this heifer. EPD's assumed she would genetically be a mathematical average of her parents. Suppose that no birth weights were recorded for any of her ancestors for several generations. That BW EPD can't be accurate, can it? But it will still be calculated with the best data available based on all the ancestors. </p><p>EPD's are most accurate when data is measured and submitted. So you start recording birth weights on her calves. First one has a lower than expected bw per the scale. The EPD calculating machine will take note of that and make an adjustment of the EPD based on what we will call a statistical model. How much adjustment? Well, she has only had one calf. Maybe it is a statistical outlier based on how those chromosomes combined. Her BW EPD accuracy will still be pretty low. Now after she has 10 calves with recorded birth weights, we and the EPD model machine will have a clearer expectation of her BW EPD and the accuracy of that EPD will be much higher. And can be trusted more.</p><p>Say there were no EPD's. Your evaluation of BW could only be based on what you see from the birth weights of the cow family and their calves. That's sort of the same way that EPD's work. The more data - the greater the accuracy and confidence in the number.</p><p>Don't expect that a yearling bull can always be counted on to match up with his EPD's. Accuracy is too low. But even then, would you rather not have the prediction? If you are going to cut hay, do you use the weather <strong>forecast </strong>even though the accuracy is not 100% or just look at the sky 4 days before you hope to bale?</p><p>My message is that EPD's with low accuracy are better than nothing, but understand that low accuracy means lower confidence in the number. High accuracy EPD's - if you don't use those, you are missing out. High accuracy comes from lots of progeny - older bulls that have been used a lot. As has been said many times, EPD's are just one tool for evaluation and selection. Don't make them the <strong>only </strong>tool but understand them to allow them to be <strong>a</strong> tool.</p></blockquote><p></p>
[QUOTE="simme, post: 1727203, member: 40418"] So where do EPD's come from and how are they calculated? Random number generator? I don't think so. It is an estimate, an expectation. It is not a promise or a guarantee. It is like a forecast based on data. Take a trait that is easily measured like birth weight. You just need a scale. Simmental has a planned mating EPD estimating tool. Just plug in the sire and the dam that you plan to mate and hit the button. It is mostly just an average of the EPD's of the sire and dam. (I say mostly because there may be some "expected" adjustment based on heterosis depending on the mating.) We know that the actual mating will combine genetics from the parents with some variation. There is no equation that calculates the birth weight of the calf from the EPD's. But if both parent families have heavier than average birth weights, isn't it logical that there is a high probability that the calf they produce will have a higher birth weight than average for the breed? But also possible that it could be low birth weight? EPD's will predict a higher expected bw. If the actual bw from the scale is lower than breed average, is the EPD wrong? Maybe or maybe not depending on environment. Let's ignore that environment thing for now. So bw epd is high and the calf is one of the lightest in the pasture. Well maybe that EPD [B]is [/B]wrong. Let's look at the accuracy. Say this is the first calf from this heifer. EPD's assumed she would genetically be a mathematical average of her parents. Suppose that no birth weights were recorded for any of her ancestors for several generations. That BW EPD can't be accurate, can it? But it will still be calculated with the best data available based on all the ancestors. EPD's are most accurate when data is measured and submitted. So you start recording birth weights on her calves. First one has a lower than expected bw per the scale. The EPD calculating machine will take note of that and make an adjustment of the EPD based on what we will call a statistical model. How much adjustment? Well, she has only had one calf. Maybe it is a statistical outlier based on how those chromosomes combined. Her BW EPD accuracy will still be pretty low. Now after she has 10 calves with recorded birth weights, we and the EPD model machine will have a clearer expectation of her BW EPD and the accuracy of that EPD will be much higher. And can be trusted more. Say there were no EPD's. Your evaluation of BW could only be based on what you see from the birth weights of the cow family and their calves. That's sort of the same way that EPD's work. The more data - the greater the accuracy and confidence in the number. Don't expect that a yearling bull can always be counted on to match up with his EPD's. Accuracy is too low. But even then, would you rather not have the prediction? If you are going to cut hay, do you use the weather [B]forecast [/B]even though the accuracy is not 100% or just look at the sky 4 days before you hope to bale? My message is that EPD's with low accuracy are better than nothing, but understand that low accuracy means lower confidence in the number. High accuracy EPD's - if you don't use those, you are missing out. High accuracy comes from lots of progeny - older bulls that have been used a lot. As has been said many times, EPD's are just one tool for evaluation and selection. Don't make them the [B]only [/B]tool but understand them to allow them to be [B]a[/B] tool. [/QUOTE]
Insert quotes…
Verification
Post reply
Forums
Cattle Boards
Breeding / Calving Issues
Are we too dependent on EPD's?
Top