Even EPDs have their place in genetic selection

By Jim Munsch Coon Valley, Wisconsin—I am often asked about the best breed for grass-based beef production. Most grass-fed experts say that a high proportion of English genetics is important. I lean that way myself — we have Angus.

But I always tell people that the specific breed is not the most important factor in grass-fed, as within each breed there are animals that can do well in your grass system, and those that won’t. When asked about breeds, I instead treat the questioner to a long explanation involving “type” and the methods for managing genetics.

That’s because regardless of your production system or what class of animals you’re selling, managing genetics is so important that you must take it into your own hands. The process of controlling genetics needs to be done with some system or combination of systems that rely on actual data ­— both from your own farm and from other sources — to help make decisions.

You need to know what kind of cow works for you in terms of production efficiency and making beef that meets the needs of your specific grass-fed or grass-based market. That cow needs to be able to do the job on your farm — not someone else’s — opening the possibility of using bulls from your own herd. In improving your herd’s performance, the cow herd requires just as much emphasis as the bull genetics being used on that herd.

And here is something that amounts to heresy in the grass-fed world: EPDs can have a place among the data you employ to make progress. When selectively used, these numbers from the world of conventional beef production can offer valuable information to someone who is truly serious about making genetic progress with a grazing-based herd.

The first thing here is to understand your goal for the cattle in your operation. What is the perfect finished animal? What is the perfect cow?

Defining the “perfect” animal depends on the cattle business you’re in. If you’re selling feeders to grass finisher, then type is important. If you sell finished animals to a consolidator or processor/wholesaler, you need to aim for what that buyer wants in terms of size, finish, ultrasound data and more. If you sell meat to customers, you should understand what they value as important, and which of these values have genetic components.

To develop those animals, I believe you need numbers. And to have numbers, you need to do some measuring. But before getting into a plan for a measurement system, it’s important to understand the performance characteristics and requirements for your place and your business.

For my own business, I like to look at this by addressing questions in two areas: What characteristics affect the value of the product for our customers, and which ones affect cost in our system of production?

For our customers I rank tenderness and having the “right” amount of marbling as very important. Both have a direct genetic component. Other genetic traits that affect tenderness and taste are fat cover and animal temperament — fat cover because it affects taste and shrink during dry aging, temperament because animal scientists tell us that it is one of a couple dozen factors determining tenderness.

For effect on cost, the top factors are the ability of a cow to get bred on time, have a live birth and raise the calf to weaning. You’d like the cow to do this for as many years as possible, and do all of it on feed that doesn’t cost much to produce. Right after that is the genetic contribution to efficient weight gain coming from both cow and bull. The cow’s ability to do these things is propped up by the following measurable traits: calving ease, good udder and feet, mothering tendencies, producing a vigorous calf, temperament, ability to perform on low-cost winter feed, and breeding ease.

At our place, we want calves to gain well after they’re on their own. Because our business is selling finished animals off of grass instead of as feeder calves, we’re not too interested in what the calf weighs at six months or weaning, but rather what they’ve done by our usual finishing time. It seems that cows are a balance between producing milk and putting on weight. A high-milking cow can wean a fat calf, but if the genetic package in the calf is for milk at the expense of weight gain, the calf won’t do well from weaning to finish. Vice-versa is also true here.

In devising our measuring system, we focus on the finished animal on one hand, and cow performance on the other.

With finished animals we grade each carcass to determine marbling and fat cover. We measure live weight, hot carcass weight and salable meat weight. We do tenderness testing. The data are separated by “contemporary group” — that is, we bunch animals by the seasons because some are finished on pasture, while others are finished largely on harvested forage. All of this data is racked up by dam and sire for indications of genetics that outperform others.

Cow data we keep besides the above include calving date relative to service date, calving ease, cleaning, body condition at the end of both the grazing season and winter, teats and udders, temperament and mothering. We keep count of years of “good conduct.” With potential sires we also do genetic testing for the genes that indicate tenderness. Ultimately, we cull dams and pick sires informed partially by this information. In very simple terms, we keep bulls and replacement heifers from dams that perform the best on these traits.

We prefer to keep bulls from our gene pool until we get too close to inbreeding. We are thus willing to give up some hybrid vigor in gaining control of genetics.

Only when inbreeding comes into view do we reach outside the herd and bring in male genetics, and then only after considerable analysis. We bring in semen only, and use it on only a small number of females to see what we get.

In my mind there are three systems for helping to determine what outside animal to pick: Breed Expected Progeny Differences (EPDs), phenotypic measurements, and observation of the production system where the animal is doing its thing in order to assess whether the system is like your own.

The EPD-based system used in the general beef industry has been ridiculed by some grass-fed industry commentators. Some of this talk is justified, but a lot is not. It isn’t so much the data, but how the data are used that should be called into question.

The underlying assumption is that the data in the EPD system come out of a grain model and therefore doesn’t apply to a grass model, and thus shouldn’t be used. However, unlike the dairy industry, the main engine of conventional cattle production — the cow herd — is predominately pasture-based. Most good beef backgrounding operations are pasture-based.

So the only suspect EPD data are around the finishing phase. I even believe from observation that breed finishing data indicate performance on grass, although I can’t prove it statistically. I’d like to see an academic or breed organization take up the challenge of proving that. Everything else being equal, I’d pick genetics that have good finishing data for marbling, backfat, yield and the like, even if the finishing was done on grain.

There is a lot of information out there. If you take the data set for the Angus breed, there is a wide range of information for sire and dam selection that lends itself to grass systems: birth weight, calving ease, weaning weight, frame score, mature size and temperament, to name a few. So when we have to go outside for semen, we use as much as we can from the Angus EPD system.

We also look at systems that rely on measurements of physical characteristics, or phenotype, as these balance EPD biases. Basically, what we do replicates the eyeball system used by cattlemen for five or six thousand years. It completes frame score with thickness. It estimates forage intake capacity and general stamina with physical dimension. It ties conformation to breeding ease. It connects rate of maturation to animal size. It gives visual analysis to breeding capability. It is, in short, a more rigorous way to “eyeball” cattle.

Finally, we are working in complex biological systems we call “farms” as we make our genetic determinations. Any investigation of outside genetics must account for our farms, including soils, climate, topography and how we manage. If you find a potential sire that checks out well in the measurement systems above, and the farm is close to being like yours, that’s the one to go for.

A word of caution: even if you find a sire or dam you think is just wonderful, this genetic package isn’t the silver bullet, as it will bring only half of the genes to any mating. Research says that the ability of those genes to influence a characteristic is generally less than 50%. These outside animals can trend you in a desired direction, but you need to measure outcome in successive generations with a system like I’ve described here.

Maybe all of this seems like too much time and effort, but it doesn’t have to be. Most of my cow and calf data go in the pocket notebook I carry around all the time. My processor collects all the weights, so the only real work on my part is to inspect and grade (which the processor has offered to do when I can’t). To me, the extra attention seems like a small price to pay for moving genetics in a way that will make me more money with the same overall farming effort.

We’ve found that as you go forward with this sort of data-driven approach, the systems converge. It is a case of form following function: good impact on progeny, good eye appeal. The animals that perform well on our place begin to produce data shown by the predominant systems as being favorable. Measuring and using the data works.

Jim Munsch has a grazing-based cow-calf operation near Coon Valley, Wisconsin.