Real World Data VS EPDS

Help Support CattleToday:

smnherf

Well-known member
Joined
Jan 24, 2006
Messages
452
Reaction score
0
Location
South Dakota
I sent all of my steers except for 3 down to a feedyard in KS. I collected data on all the steers both commercial and the ones out of registered cows.

My old Mac 0232 bulls data was very interesting. He had 9 steers in the bunch, mostly out of first calf heifers. He has epds of only 64 on YW and .15 REA and -.13 IMF which would rank him in the bottom 3% for IMF of the breed. But his steers outgained the pen average and 7 of the 9 steers graded choice. The pen was comprised of a Boudler son I own, and AI sires such as M326, 19D, H33, 24F, and Predestined. All of these have much higher epds than 0232.

I have to admit that his epd profile has held me up on how much I have used the bull, but judging from the real world data, its apparent the epds aren't accurate for him. I am curious how often this happens with the epds not matching the ultrasound data. Anybody on this board have any experience with this too?

Brian
 
A lot of the discrepancy has to do with #1 the cows but more importantly the accuracy of the EPD. Until they reach around 80% they're kind of a a SWAG in many cases.
 
It seems to me that the epd's are taking bigger swings from analysis to analysis on some individuals than they used to. I suspect this is because of a greater emphasis on recent data in the analysis as opposed to more emphasis on parentage data...just my WAG. I also suspect this is allowing some individuals to manipulate the data more easily and "build" epds. As for the ultrasound data, my first tip off that something wasn't adding up was when I had so many people telling me that you couldn't select for rideye just by looking at the animal....in other words the ultrasound was showing more ribeye on animals that appeared to have less ribeye. The problem withthat theory is our universities had been training judgeing teams to do visual rating of ribeyes for years and slaughtering the animal to confirm the visual appraisals. The ultrasound work I had done in my own herd seem to identify larger ribeyes on animals that visually appeared to have larger ribeyes, so within herd the ultrasound seemed pretty accurate. Who knows what is going on in other herds, when people tell me you can't visually rate larger ribeye cattle, I know someone is off base somewhere. I haven't done any ultrasound work on my cattle in a few years now.
 
alexfarms":irqsppez said:
As for the ultrasound data, my first tip off that something wasn't adding up was when I had so many people telling me that you couldn't select for rideye just by looking at the animal....in other words the ultrasound was showing more ribeye on animals that appeared to have less ribeye. The problem withthat theory is our universities had been training judgeing teams to do visual rating of ribeyes for years and slaughtering the animal to confirm the visual appraisals. The ultrasound work I had done in my own herd seem to identify larger ribeyes on animals that visually appeared to have larger ribeyes, so within herd the ultrasound seemed pretty accurate. Who knows what is going on in other herds, when people tell me you can't visually rate larger ribeye cattle, I know someone is off base somewhere. I haven't done any ultrasound work on my cattle in a few years now.
You can;t tell just by external visual appearance. Usually those that appear to have a large rib eye will, usually! But it isn;t alwasy what it appears.
We've retained ownership and got kill data back on enough others to know that what you see isn;t always what you get. I sure wouldn;t base a bull selection strictly on visual appraisal if I was looking to improve RE.
 
I sure wouldn;t base a bull selection strictly on visual appraisal if I was looking to improve RE.

I won't trust EPDs or scan data unless I can also visually see the strong loin development.

What people seem to forget is area = height x width. What the REA may be lacking in thickness it may make up width or vice versa.

All that said I don't want to breed flat muscled cattle so even though the animal may scan very well it needs to meet my visual standards for saleability.
 
KNERSIE":m3v5p3uh said:
I sure wouldn;t base a bull selection strictly on visual appraisal if I was looking to improve RE.

I won't trust EPDs or scan data unless I can also visually see the strong loin development.

What people seem to forget is area = height x width. What the REA may be lacking in thickness it may make up width or vice versa.

All that said I don't want to breed flat muscled cattle so even though the animal may scan very well it needs to meet my visual standards for saleability.

Your right, My old college judging coach when teaching us to judge hogs was they had to have base width and that muscle has shape whereas fat doesn't. Understanding that principle is helpful in judging cattle too. Fat can make cattle look a lot thicker than they really are. Looks alone can't determine REA. A local large Chi angus breeder made the comment a few years ago after getting into the carcass traits that they are realizing that cattle don't have to be 3 axe handles wide in order to have a good carcass. That being said though I do beleive that REA is probably a lot easier trait to measure through ultrasound than what IMF is as management plays a larger role in imf scores.

I am to the point that I won't use bulls that don't have scan data on them and their mgs and they need to have some actuall data on their offspring before I use them in any volume.

Brian
 
It gets more interesting as I look more into the data.

I have a Boulder son here that has 26 scans all from my own herd with a top 35% ranking in the breed for IMF epd. Their is a fair amount of data collected on his sire and half sibs in other herds. He had 7 calves in the group and 2 graded choice. 3 of the 7 of these were out of half sisters to 0232 and 2 more of them had 0232's sire as their mgs.

Now on 0232's calves, 5 of the 9 had his sire as the maternal grandsire. So they were fairly tightly bred back to the old Mac 117D bull but still graded choice, whereas when bred to the Boulder son, they reduced the % choice.

Brian
 
smnherf":2siryk55 said:
It gets more interesting as I look more into the data.

I have a Boulder son here that has 26 scans all from my own herd with a top 35% ranking in the breed for IMF epd. Their is a fair amount of data collected on his sire and half sibs in other herds. He had 7 calves in the group and 2 graded choice. 3 of the 7 of these were out of half sisters to 0232 and 2 more of them had 0232's sire as their mgs.

Now on 0232's calves, 5 of the 9 had his sire as the maternal grandsire. So they were fairly tightly bred back to the old Mac 117D bull but still graded choice, whereas when bred to the Boulder son, they reduced the % choice.

Brian

Since you are using your steers as examples, give us a little more information. The two biggest indicators of IMF score(quality grade) are age(days on feed) and back fat thickness. How do these two factors correlate to how your calves ranked in quality grade?

Do you realize that you are, on one hand, praising the importance of untrasound data and their associated epds and on the other hand giving examples that discredit the ultrasound epds? Sometimes if is hard to grasp the obvious isn't it?
 
KNERSIE":2khcatkb said:
Do you have a photo of Mac 117D?

23819454_L.JPG


This is the only picture I have of him.
 
alexfarms":2srmnogi said:
smnherf":2srmnogi said:
It gets more interesting as I look more into the data.

I have a Boulder son here that has 26 scans all from my own herd with a top 35% ranking in the breed for IMF epd. Their is a fair amount of data collected on his sire and half sibs in other herds. He had 7 calves in the group and 2 graded choice. 3 of the 7 of these were out of half sisters to 0232 and 2 more of them had 0232's sire as their mgs.

Now on 0232's calves, 5 of the 9 had his sire as the maternal grandsire. So they were fairly tightly bred back to the old Mac 117D bull but still graded choice, whereas when bred to the Boulder son, they reduced the % choice.

Brian

Since you are using your steers as examples, give us a little more information. The two biggest indicators of IMF score(quality grade) are age(days on feed) and back fat thickness. How do these two factors correlate to how your calves ranked in quality grade?

Do you realize that you are, on one hand, praising the importance of untrasound data and their associated epds and on the other hand giving examples that discredit the ultrasound epds? Sometimes if is hard to grasp the obvious isn't it?

I realize exactly what I am saying. I use the data available. I prefer actuall kill data, but if that isn't available, ultrasound data is usefull if their is enough data collected in several herds. My overall question though is how often does ultrasound miss with respect to IMF?

The calves were all born from about Mar 10th to May 10th. The 0232's would tend to be about 30 days older as I breed my heifers to calve earlier than my cows. Backfat on 0232's would be .51 ave and ave bf on boulder would have been .45 with 4 of the five being Select yg 2's. 1 of 0232's select was a YG 2 also. They were all weaned, backgrounded and finished in the same contemporary group though.
 
smnherf":2i9wzxzv said:
I have to admit that his epd profile has held me up on how much I have used the bull

this statement has boggled my mind about this business for as long as i can remember. what is the deep down gut reason for the hesitation in using the bull for an epd profile.
 
smnherf":37h7r8gu said:
alexfarms":37h7r8gu said:
smnherf":37h7r8gu said:
It gets more interesting as I look more into the data.

I have a Boulder son here that has 26 scans all from my own herd with a top 35% ranking in the breed for IMF epd. Their is a fair amount of data collected on his sire and half sibs in other herds. He had 7 calves in the group and 2 graded choice. 3 of the 7 of these were out of half sisters to 0232 and 2 more of them had 0232's sire as their mgs.

Now on 0232's calves, 5 of the 9 had his sire as the maternal grandsire. So they were fairly tightly bred back to the old Mac 117D bull but still graded choice, whereas when bred to the Boulder son, they reduced the % choice.

Brian

Since you are using your steers as examples, give us a little more information. The two biggest indicators of IMF score(quality grade) are age(days on feed) and back fat thickness. How do these two factors correlate to how your calves ranked in quality grade?

Do you realize that you are, on one hand, praising the importance of untrasound data and their associated epds and on the other hand giving examples that discredit the ultrasound epds? Sometimes if is hard to grasp the obvious isn't it?

I realize exactly what I am saying. I use the data available. I prefer actuall kill data, but if that isn't available, ultrasound data is usefull if their is enough data collected in several herds. My overall question though is how often does ultrasound miss with respect to IMF?

The calves were all born from about Mar 10th to May 10th. The 0232's would tend to be about 30 days older as I breed my heifers to calve earlier than my cows. Backfat on 0232's would be .51 ave and ave bf on boulder would have been .45 with 4 of the five being Select yg 2's. 1 of 0232's select was a YG 2 also. They were all weaned, backgrounded and finished in the same contemporary group though.

Common sense would suggest to me that there are a few environmental affects which are playing in 0232's favor which may have helped give his calves an advantage in IMF. The calves are older, the calves carry more backfat and the calves are out of first calf heifers(a possible advantage). Being out of first calf heifers may have given the calves less overall growth potential as compared to calves out of mature herd mates. With less overall growth potential, frame score may have been reduced which could have increased feed efficiency(once on feed) which means they could have started laying on more fat more quickly and been farther along at a set endpoint.
How often does ultrasound miss the mark with respect to IMF? I think it has been very usefull for feedyards to use ultrasound to determine an optimum marketing point. Any use beyond that looks to me to be pretty fuzzy, there are just so many environmental factors that go into carcass traits that it questionable whether they are being properly measured or adjusted. I wouldn't let ultrasound carcass epds be a deciding factor in a selection decision. JMHO
 
Being out of first calf heifers may have given the calves less overall growth potential as compared to calves out of mature herd mates.

Being out of first calvers would have no effect on growth POTENTIAL, but it may affect the growth environment.
 
Hereford76":3655dvyh said:
smnherf":3655dvyh said:
I have to admit that his epd profile has held me up on how much I have used the bull

this statement has boggled my mind about this business for as long as i can remember. what is the deep down gut reason for the hesitation in using the bull for an epd profile.

I won't use a bull just for his epd profile. He needs to have more than just the right profile for me. His epd profile may exclude him from use in my herd though. My statement was based on his lack of IMF epd. I try to avoid extremes on both ends and he is on the low end.

He was one of the first 10 bulls I ever sold. I origninally sold the bull to a neighbor who used him for a couple years on commercial angus heifers and then for a couple years on his cow herd. Meanwhile, I had his daddy here and used him several more years. He had a high percentage of choice carcasses on his steers. When the neighbor was done with him I bought him back because I thought he was one of the best bulls I raised and many of the AI sires I used hadn't lived up to their billing. Based on what his cow family had done I decided to find a way to use him even if I had to inbreed him some. Now understand, he is related to probably 75% of my cow herd both purebred and commercial. His full brother whom I used 1 year clean up after AI didn't scan very well for IMF either so my concern was that his calves wouldn't be as easy for me to market as better epd cattle would be. I am new to the bull business and trying to establish a customer base and having better epds for a marketing tool is helpfull. In this case, epds have been to my detriment as several of my first customers where watching epds carefully as they were retaining ownership on their calves.
 
alexfarms":1h9u401k said:
Common sense would suggest to me that there are a few environmental affects which are playing in 0232's favor which may have helped give his calves an advantage in IMF. The calves are older, the calves carry more backfat and the calves are out of first calf heifers(a possible advantage). Being out of first calf heifers may have given the calves less overall growth potential as compared to calves out of mature herd mates. With less overall growth potential, frame score may have been reduced which could have increased feed efficiency(once on feed) which means they could have started laying on more fat more quickly and been farther along at a set endpoint.
How often does ultrasound miss the mark with respect to IMF? I think it has been very usefull for feedyards to use ultrasound to determine an optimum marketing point. Any use beyond that looks to me to be pretty fuzzy, there are just so many environmental factors that go into carcass traits that it questionable whether they are being properly measured or adjusted. I wouldn't let ultrasound carcass epds be a deciding factor in a selection decision. JMHO

Maybe they did have an advantage, maybe they didn't, I don't know. Being out of first calf heifers may have been a disadvantage too. IMF is determined early on and the fact that they nursed 2 yr old heifers late into the fall on low protein grass shouldn't be to their advantage either versus older cows that can keep lactating longer.

I dont see how you can make any assumptions on growth potential or frame though. It is possible that they were fatter becasuce of the difference in ADG too. The 0232's gained 4.04 per day with a carcass wt of 851 vs 3.72 per day and 803 lb CW for the Boulder son. Sure they are smaller framed but are they more efficient, I don't know. Without individual intake data, and I have some idea of which sire group ate more based on past tests, all I can do is guess at that. It is possible that the Boulders with more days on feed would have a higher % grading choice and I suppose that the EPD model would adjust several of the IMF scores up to get to a standard level for bf, but then that final epd number won't take into account a very important component of feedlot profitability would it, and that is days on feed and ADG. The 0232's would have a huge advantage then in profitability and it wouldn't show in any of the EPDs. Is this logical thinking?

Brian
 
smnherf":sea2ancl said:
I sent all of my steers except for 3 down to a feedyard in KS. I collected data on all the steers both commercial and the ones out of registered cows.

My old Mac 0232 bulls data was very interesting. He had 9 steers in the bunch, mostly out of first calf heifers. He has epds of only 64 on YW and .15 REA and -.13 IMF which would rank him in the bottom 3% for IMF of the breed. But his steers outgained the pen average and 7 of the 9 steers graded choice. The pen was comprised of a Boudler son I own, and AI sires such as M326, 19D, H33, 24F, and Predestined. All of these have much higher epds than 0232.

I have to admit that his epd profile has held me up on how much I have used the bull, but judging from the real world data, its apparent the epds aren't accurate for him. I am curious how often this happens with the epds not matching the ultrasound data. Anybody on this board have any experience with this too?

Brian


Brian...without studying all your data, my first thought is the Mac 0232 bull produced more heterosis (illogical skew) with the steer dams even though 70% of your cows have common genetics to 0232. If you continued over a number of years with a different mix of genes, the higher EPD bulls should prove superior to your 0232 results in the categories you are expecting.
 
I don't care for EPD's, although I will say this - if we were going to judge 117D's Ribeye by that picture, he looks pretty damn good. Long bodied with a bulging ribeye = lots of meat where it's worth $$$. I'd use him in a heartbeat. Some baldies out of him and our GallowayX cows would be a fancy combination.
 
smnherf":24ytf851 said:
alexfarms":24ytf851 said:
Common sense would suggest to me that there are a few environmental affects which are playing in 0232's favor which may have helped give his calves an advantage in IMF. The calves are older, the calves carry more backfat and the calves are out of first calf heifers(a possible advantage). Being out of first calf heifers may have given the calves less overall growth potential as compared to calves out of mature herd mates. With less overall growth potential, frame score may have been reduced which could have increased feed efficiency(once on feed) which means they could have started laying on more fat more quickly and been farther along at a set endpoint.
How often does ultrasound miss the mark with respect to IMF? I think it has been very usefull for feedyards to use ultrasound to determine an optimum marketing point. Any use beyond that looks to me to be pretty fuzzy, there are just so many environmental factors that go into carcass traits that it questionable whether they are being properly measured or adjusted. I wouldn't let ultrasound carcass epds be a deciding factor in a selection decision. JMHO

Maybe they did have an advantage, maybe they didn't, I don't know. Being out of first calf heifers may have been a disadvantage too. IMF is determined early on and the fact that they nursed 2 yr old heifers late into the fall on low protein grass shouldn't be to their advantage either versus older cows that can keep lactating longer.

I dont see how you can make any assumptions on growth potential or frame though. It is possible that they were fatter becasuce of the difference in ADG too. The 0232's gained 4.04 per day with a carcass wt of 851 vs 3.72 per day and 803 lb CW for the Boulder son. Sure they are smaller framed but are they more efficient, I don't know. Without individual intake data, and I have some idea of which sire group ate more based on past tests, all I can do is guess at that. It is possible that the Boulders with more days on feed would have a higher % grading choice and I suppose that the EPD model would adjust several of the IMF scores up to get to a standard level for bf, but then that final epd number won't take into account a very important component of feedlot profitability would it, and that is days on feed and ADG. The 0232's would have a huge advantage then in profitability and it wouldn't show in any of the EPDs. Is this logical thinking?

Brian

Yes, I think you are being very logical. I think that is why you asked the questions in the first place. In my opinion there are alot of variables that could distort any conclusions anyone would try to make from one of these tests. Comparing raw (unadjusted) data would be a mistake and trying to satisfy all of the potential adjustments would be virtually impossible. I will stand by my "assumptions" on lowered growth potential in calves from first calf heifers as compared to their mature herd mates and the reason is that the environmental differences would begin during gestation. That is why 24 month old heifers have their offspring's birth weights adjusted up. That one factor will never be overcome for calves out of first calf heifers. Their genetic potential wouldn't be affected though, of course.

THE conclusion you can draw about your 0232 bull is he is capable of siring calves that will hit targets in a feedlot and that is something to be proud of. I remember when these ultrasound epd's first came out and one of the points the originators made was: "This technology will enable us to identify and eliminate those cattle that just can't develop the necessary carcass traits to perform in a feedlot." Your 0232 bull definately doesn't need to be culled for lack of desirable carcass traits.
 

Latest posts

Top