Plots Suck. Part II.

Part I of this series showed that having confidence in the results of a plot matters. It’s asking an awful lot of a 5 or 10-acre strip plot to represent hybrid selection across hundreds or thousands of acres. When seed companies use this method they rely on dozens to hundreds of them to understand how different environments impact hybrid performance. This is commonly referred to as GxE, or genetics by environment. But not every grower manages their crops the same so having different locations introduces another factor – management (M). Now we have a GxExM interaction and you can quickly see how the whole picture of product performance can get muddled (e.g. one grower may apply fungicide or late season nitrogen while another doesn’t). That’s why quality data for an individual plot are critical. The data are important to you for your farm but also think of it as a data link in a decision chain.

Are there other ways of evaluating hybrids? Sure. I personally prefer large strips of 5 or more acres for each individual hybrid being tested. But, the yield has to be checked with a weigh wagon or a calibrated yield monitoring system. Using yield maps is fine as long as the data are of good quality. So what are sources of error that can impact plot results? Here are a few:

Harvest ruts can impact next year’s yield but what happens when a plot entry is planted in this affected area while the others are not?

  • Hybrids planted in the pinch rows (see Plots Suck, Part I).
  • Plot planted last, long after most acres were planted.
  • Placing the plot in a part of the field that results in an entry being planted in an undesirable spot that clearly impacts yield (e.g. an entry happens to be planted in a low spot).
  • Supplied seed has different seed sizes and no adjustments are made resulting in different stands across the plot.
  • Sprayer runs over or “leans over” some of the rows of a single entry but not the same number across all entries in the plot.

It’s not possible to eliminate every source of error. Remember, the goal is to have each entry in the plot treated equally and to walk away at the end of harvest with a high degree of confidence that the results are correct. When we achieve this we can say, “that plot didn’t suck.”

I want y’all to do well. God bless our farmers!

Plots Suck. Part I.

Think about this: You farm 4,000 acres and you need a way to figure out which hybrids perform best. So what do you do? You pick 1 field and plant 1 plot. Yep, a 5-acre plot represents 4,000 acres. That’s all the time you have for this. Seriously, that’s what most of you do. The rest of you ask your neighbor what worked on their farm. And it doesn’t matter that they farm differently than you; rather, they’ve survived this long so they must be making the right decisions on hybrid selection. And, then, next year when the hybrid you chose wasn’t as good as the previous year or as good as the neighbor was bragging about, you get upset. And call your seed rep who then calls his agronomist who then rolls his eyes wondering why we repeat this cycle. Plots suck.

As an agronomist in the business for over 20 years, I estimate I’ve been involved with about 8,000 plots. Of these, 7,500 were a complete waste of time. Exaggeration? Not much. Seriously? Yes, seriously.

While it takes up more space, split the planter between two hybrids. And while it may take more time to plant, the outcome is better and you get a whole lot more free seed.

The goal of plot work is to evaluate product performance with a high degree of confidence. Confidence can have a statistical connotation but at the end of the year you want to walk away knowing that the test was fair and every entry had an equal chance of performing to its highest potential. So let’s explore this.

You have 12 hybrids and a 24-row planter and very little time to plant the plot that the sales rep has been pestering you about for the last month. (The reps don’t haven’t much of a life – all they do is drive around dropping off plot bags in the hopes that somebody will feel sorry for them and plant the damn things.) You’re under pressure and won’t devote more than an hour or so to get this done. What do a lot of growers do in this setting? They fill the planter with 6 hybrids, 4 rows each, refill at the end, plant back and they’re done. In this scenario, 4 of the 6 hybrids in the pass are treated equally, the other 2 are screwed. The 2 that were treated unfairly (those under the tractor tires and in the pinch rows) will likely yield less leading you to draw the conclusion that they’re not as good. Consequently you don’t order a single bag of either and tell the neighbors that those hybrids suck. You know what? The plot sucked.

I am fond of saying “to measure inaccurately is to not measure at all.” I firmly believe this. Think about this scenario – many growers will take the time to plant a quality plot but won’t take the time to use a weigh wagon. They harvest the plot using a combine with a yield monitor that hasn’t been calibrated and surely won’t be calibrated with each change in hybrid. This is another scenario where we walk away with limited confidence in the outcome of the plot. We justify this by telling ourselves that, if there is a source of error in the uncalibrated system, it affects each entry equally and in the same way. I am here to tell you that this is not a good assumption. Which reminds me, I’ve never heard a grower say, “if I get some time I’ll send you harvest maps of the plot where we used an uncalibrated combine.” Discomforting, to say the least.

Plots can offer a lot about product behavior and performance. To the best of your ability plant them when most other acres are being planted (certainly not last with a planting date that is not typical for the area). Plant them with the same care and attention as you give all the other acres and respect the time and effort that the breeders have invested in creating a bin-busting hybrid that can yield 500+ bu/a. Visit with the agronomists and sales reps beforehand to design a quality plot experience from start to finish.

I’ll leave you with this – if you put out a quality plot and my hybrids don’t perform well, I will respect that. If put out a plot that everybody and their dog knows wan’t done well, we’ll all talk about you every day at the local coffee shop and how your plot sucked. Don’t be that person.

I want y’all to do well. God bless my farmers!