USDA said in September that Indiana and Ohio would harvest corn yields well below normal, more so in Indiana than Ohio. The reports are usually adjusted in the next few reports that followed. Some were betting on still lower estimates, some on higher.
Anyone who sampled the 2012 Cropwatch field carefully would wonder where the 100 or 100-plus bushel per acre estimates for state figures were coming from. The field averaged out to 40 bushels per acre, considering several samples from all five soil types in the field. Again, that is just an estimate.
Here are factors at play that could factor into future reports.
Acreage harvested – The August USDA estimate made little or no adjustment for failed crop acreage. If acreage is dropped in future reports, that could actually push yields slightly higher on acres that are harvested.
Varying reports – So-called professionals have been all over the board since then, some claiming national yield as high as 140 bushels per acre, others national yield as low as 114 bushels per acre. Obviously, different professionals are using different tools or differing crop models to come up with yield estimates.
Shallow kernels – The normal formula for estimating yield before harvest divides the total number of kernels by 90. Some suggest that it should be divided by at least 100 this year, which makes the resulting estimate smaller, since kernels aren't as deep as in normal years. Some agronomists who have reported so far say fields are running up to 20% less yield than they estimated before harvest.
Historical trends – There are about as many years that yields go up from August estimates on to final yield as they go down. However, after hot, dry seasons during pollination, the trend is for USDA to drop corn estimates after August.
Far-out theory – Some believe USDA may have held yields up rather than shock the market with even more realistic, lower numbers. Again, it's a theory, and not founded in fact, or not a swipe at USDA.
Estimating methods – This is fact. Normally when USDA boots on the ground gather data for August estimates, ears aren't formed yet. This year in some cases they were. If they could already measure small ears or count barren stalks, they may have already built in adjustments that would normally be built in later.
Hot nights theory – Although it got lost in the shuffle this year because drought and heat was so overwhelming, nighttime temperatures during peak pollination and early grain fill were higher than normal. That's been the case the past two seasons. Some believe it's part of the reason why corn estimates were lowered in each of the last two seasons after August, especially in 2010.