USDA Got the Corn Numbers Wrong

Posted October 26, 2026 · Updated October 31, 2026

Update, October 31. NASS put out a second notice on October 24 that settles two things this page had marked as unclear. The cause: NASS says the wrong acreage weights were used when it combined irrigated and non-irrigated objective-yield samples for Kansas and Nebraska. The Kansas number: the corrected September yield is 118 bushels, down 3. The same day, USDA and NASA signed an agreement on satellite data for crop estimates, covered further down.

On October 23 USDA’s National Agricultural Statistics Service corrected its September corn data after finding a processing error. Kansas and Nebraska yields moved, the national September figure moved with them, and the October forecast was left standing. Colorado was not among the corrected states — which is not the same thing as Colorado being unaffected.

This is worth a page for two reasons, and neither of them is the market. One is what the error appears to have been about. The other is which decisions are about to be made using the same machinery.

What Was Corrected

NASS says the error reached September corn plant populations, ear counts, yields and production for Kansas and Nebraska, the ten-state averages for plant populations and ear counts, and the national corn yield and production forecasts. The corrected figures are in the Crop Production reports and in Quick Stats.

That last point is where the argument is. An error large enough to move two state estimates and the national figure was found in September, and the October forecast built on the same survey came through untouched. Analysts have said plainly that they do not find that reassuring, and at least one has argued the error means the corn estimate was too low rather than too high.

The Part That Should Interest Colorado

When this page first went up, coverage of the cause did not agree, and we said so. NASS has since answered the question itself. In its October 24 notice the agency says the September error came from incorrect acreage weights used to combine irrigated and non-irrigated objective-yield observations in Kansas and Nebraska. In plain terms, the dryland and irrigated samples were each counted, but they were blended in the wrong proportions.

The trade press has also reported the sample mix, and it fits that explanation. In September, roughly 90 percent of the Nebraska fields sampled were non-irrigated and about 10 percent irrigated. In October the balance came back to something ordinary — about 55 percent irrigated and 45 percent non-irrigated. A survey that saw a state as nine-tenths dryland one month and roughly half irrigated the next was not describing the same crop both times.

Colorado growers have heard this song before. Our transitional yields sat below comparable ground in other states until this association carried the problem to the SWCA board and the council took it to RMA. Our PLC program yields run under the national average because so much of our corn is dryland, which is why we keep warning members that national per-acre averages overstate what Colorado base will pay. The irrigated and dryland split is not a footnote in this state. It is the whole shape of the crop, and when a federal data system gets that weighting wrong in Nebraska, it is reasonable to ask how it handles it here.

Why It Matters in November, Not Today

The market reaction has been modest, and an agricultural economist quoted on the correction said as much — little market impact, but real damage to a trust that USDA was already working to rebuild. We would not change a marketing decision on the strength of this.

Here is the part nobody is saying out loud. The Harvest Fuel Relief payment formula runs on this data. USDA said the $8 billion will be allocated using a national loss average built from FSA planted acres, Economic Research Service cost-of-production estimates, and WASDE yield and price estimates. Those WASDE yields come out of the same NASS pipeline that just produced a correction.

Title I payment rates are expected in November. So the sequence is: a data system produces an error in September, acknowledges it in late October, and sets $8 billion in payment rates in November. That is not an accusation. It is a reason for growers to read the rate announcement closely rather than assume the arithmetic behind it is beyond question.

USDA Brings in NASA

On October 24, Secretary Brooke Rollins and NASA Administrator Jared Isaacman signed a memorandum of understanding that takes effect immediately. It commits the two agencies to work together on satellite observation, geospatial data and statistical methods for USDA’s agricultural statistics. It sits under the Data Modernization Plan USDA announced on September 1. USDA lists three areas of work: matching NASA missions and data to USDA’s statistical needs, getting NASA data into USDA’s systems at scale, and joint technical reviews of the methods behind crop condition, yield, planted and harvested area, and disaster assessments.

Of those three, the last is the one that matters on Colorado ground. A review of how yield and area are estimated is exactly where the irrigated and dryland weighting lives, so it is where the ask below belongs. A state that is split down the middle needs estimates that hold up in a dry year on dryland, not only in a good year on pivots.

USDA’s announcement gives no timeline or first deliverable, so for now this is a framework to watch rather than a change to anything you will see in a report.

What We Want

USDA has said it will add pre-publication checks to verify acreage weights and validate objective yield calculations, and that it is looking at further quality-control changes. That is the right fix and it should have existed already.

The National Corn Growers Association, which this association belongs to, has called the episode a major data processing error and pressed USDA on it. We support that, and we would add one thing from where we sit: publish the irrigated and non-irrigated breakdown alongside the state estimates. If the weighting between them can move a state yield by five bushels, growers in a state that is split down the middle have a right to see it rather than infer it.

Accurate numbers are not a bureaucratic nicety here. They set the reference prices, the benchmark revenues and now the fuel relief rates that decide what reaches Colorado ground.

Harvest Fuel Relief: what is known and what is not →
The December 11 ARC/PLC enrollment deadline →

Corrected yields and the cause of the error are from NASS’s own Agricultural Statistics Board notices of October 23 and October 24, 2026, both read directly on October 31. Where trade coverage gave a different Kansas figure, we use NASS’s. The Nebraska sample mix percentages come from trade press and are not in the NASS notices. The USDA–NASA agreement is described from USDA’s October 24, 2026 release. Anyone relying on these numbers should check NASS Quick Stats and the Crop Production report directly. CCGA administers no USDA program and produces no yield estimates.