National Ag Statistic's Service didn't appear to learn anything from their near-record miss on the August estimate one year ago. At least some farmers expected a change to account for certain environmental conditions that had occurred by August 1, the date the estimate was based on, but apparently it didn't happen.
“They basically used the same method they have used for a long time,” says Greg Preston, long-time head of ag statistics for USDA in Indiana. Reportedly, they have enough confidence in the method that they didn't believe it needed to be adjusted, even after a horribly inaccurate guess a year ago. Yields dropped from every estimate after the August estimate last year.
The only difference this time was that based on what his field enumerators were seeing, Preston and his staff added recommendations with their raw numbers, and he believes it may have led the statisticians a the national level to pull Indiana's projected yield, at 150 bushels per acre, back a bushel or two from what it might have been.
The problem, Preston says, is that in the August estimate, the crew of people hired to hit pre-determined spots across the state primarily count plants. Ears aren't often formed yet, and that was especially true this year. So without factoring gin other environmental factors, the numbers become based on population and little else. USDA assumes normal weather conditions for the rest of the season, which in this case would be from August 1 on.
What farmers want to know is that if USDA is trying to be accurate, and both farmers and companies and investors who buy product are using the information, why didn't they try to adjust how they do their estimates to make up for the mistake they made a year ago? Instead it was completely ignored in the report. In fact, the estimated August yield in Indiana at 150 bushels per acre was compared to 157 bushels per acre corn in 210, which doesn't sound like a horribly large drop. The trick is that the 157 is the final yield. In August in 2010 the Indiana yield was over 166 bushels per acre. Comparing 166 to 150 is about a 10% drop in yield from the August estimate a year ago.