Sep 15, 2014
Home| Tools| Events| Blogs| Discussions Sign UpLogin

From the Editor

RSS By: Brian Grete, Pro Farmer

Pro Farmer Editor Brian Grete takes time to talk with Pro Farmer Members about some of the key issues in each week's Pro Farmer newsletter.

We've got to 'trust the numbers'

Nov 08, 2013

Hello Pro Farmer Members!

I'll admit... I felt sick this morning when USDA put the national average corn yield at 160.4 bu. per acre. Not because the yield came in well above our Aug. 23 estimate of 154.1 bu. per acre, but because I've got a suspicion where the national average yield for this year will end up. And that's at 161.5 bu. per acre for a national average yield.

The other thing that frustrates me about the current USDA yield is that the number was right in front of us back on Aug. 22 when we put together our crop estimates following the Pro Farmer Midwest Crop Tour.

Many of you might remember that we changed the yield calculation used on the Crop Tour a few years back. We did that because the "new" (which we refer to as the "standard") calculation was, on average, closer to reality in estimating the state-by-state yield pegs. And, just as importantly, when all the corn yield samples are dumped into one spreadsheet (about 1,340 samples in 2013), the average "all-sample" yield is above the national average yield for any given year. And when pulling samples from Ohio, Indiana, Illinois, Iowa, Minnesota, South Dakota and Nebraska, the average yield calculated should be above USDA's national average yield.

The "old" yield calculation (which we refer to as the "adjusted" calculation) always generated yield estimates that appeared too low. And, in reality, they were too low. When pulling samples from those seven states, the average all-sample yield should not be below the eventual national average yield. (As a side note, the "adjusted" yield calculation attempts to adjust kernel size based on the number of kernel rows around the ear.)

You've probably already noticed my frustration - the "standard" calculation is typically above the eventual national average yield... the "adjusted" calculation is typically below the eventual national average yield. Therefore, the average of the "standard" and "adjusted" calculated yields should give us something pretty close to the eventual average yield. It doesn't always work... nothing "always" works when estimating a national average corn yield in the third week of August. But, it works pretty good.

Below is a table of how the average of the "standard" and "adjusted" all-sample calculated average yield ("Avg. Crop Tour Yield) compares to USDA's August 1 estimates and final estimates.

Corn Yield
Avg. Crop
Tour Yield
USDA Aug. 1
USDA Final
Avg. Crop Tour missed by

2006: Average all-sample yield was 3.6 bu. too low, but it pointed the yield in the right direction from USDA's August 1 estimate.

2007: Again, it pointed the yield in the right direction from USDA's Aug. 1 yield and ended up just 0.7 bu. above USDA's final yield estimate.

2008: A good performance... missed it by just 0.4 bu. per acre and pointed the trend in the right direction from USDA's Aug. 1 estimate.

2009: Missed the final yield by 3 bu., but pointed the trend in the right direction from USDA's Aug. 1 yield estimate.

2010: Poor finish on this crop... the Crop Tour all-sample average was too high by 6.8 bu. per acre. But it worked as a big warning that USDA's Aug. 1 estimate was way too high as the average of the standard and adjusted calculated yields was 5.4 bu. below USDA's Aug. 1 estimate.

2011: We saw the same crop USDA did in 2011... the average of the standard and adjusted yield calculations was just 1 bu. below USDA's Aug. 1 estimate and it pointed the trend in the right direction and ended up 4.8 bu. above the final yield from USDA.

2012: See what happens when you use this process on the most mature crop we've ever seen in 21 years of running the Crop Tour? That's scary.

2013: The average of the standard and adjusted yield calculation was 7.1 bu. above USDA's Aug. 1 estimate. As of this morning's report, the average of the standard and adjusted yield calculation is 1.1 bu. above USDA's current yield estimate.

That's where my frustration lies. We "had" the national average yield, but couldn't believe the numbers. We were looking at what was the most immature northern corn crop we've ever seen in 21 years of Crop Tour and "we just knew" there was no way it would finish strong.

And then the crop got the heat it evidently needed in the last week of August and first two weeks of September.

And then it didn't frost until Oct. 20.

And then the Crop Tour data worked... again.

The only good news is we just shortened up the debate that will take place in estimating the 2014 national average corn yield.

Another frustration from this morning's Crop Production Report was that our August estimates on harvested acres was better than our September estimate of harvested acres. Our Aug. 23 corn crop estimate included a harvested acreage tally that was 1.8 million acres below USDA's harvested acreage tally... we adjusted that to a 1.5 million-acre cut after viewing FSA's certified acreage data in September... USDA cut 1.9 million acres from the harvested acreage tally today. On soybeans, we cut 800,000 acres from the harvested tally back in August, then revised it to a 417,000-acre cut... and today USDA lopped 690,000 acres from harvested bean acres.


That's it for now...

... time to head home and think about this some more. (Trust the numbers... TRUST the numbers... TRUST THE NUMBERS!)

Follow me on Twitter at @ChipFlory

To join Pro Farmer, click here!

Log In or Sign Up to comment


No comments have been posted, be the first one to comment.
The Home Page of Agriculture
© 2014 Farm Journal, Inc. All Rights Reserved|Web site design and development by|Site Map|Privacy Policy|Terms & Conditions