Don, thanks for pointing out how important growing season temperature will be this coming year. And Zen, thank you for providing the data about temperature and precipitation anomalies. As I suspected it is a little easier to see how temperature and precipitation interact to influence yield when you look on a state level.
As seen in chart 4, temperature anomaly accounted for 56% of the yield variation. Similarly, taken alone precipitation anomaly accounts some 26% of the yield variation (not shown) but that's a little deceptive. If you look at the scatter plot what you see is that yield deviations are not so sensitive to moisture in a fairly wide range of values between - 1 inch and +4 inches. But too little rain or too much starts to depress yields so a polynomial is probably a better model (chart 3). The timing of the rain is probably really important as well but I was able to check that.
If you run a multiple regression analysis using temperature anomaly and a quadratic precipitation anomaly model then together they account for 71% of the yield deviation. A 3-D plot of yield deviation vs temperature anomaly and precipitation anomaly suggests how temperature and precipitation interact to influence yield (chart 1, note: the temperature & precipation numbers have been multiplied by ten in order to scale the graph). It's obvious that "Hot & Dry" is bad, "Cool & Wet" is Ok but cool & enough is better. The "Goldylocks" principle, namely "Just Right" applies.
The really important observation here (and one that I didn't fully appreciate) is "Just how critical growing season temperature will be in the coming months, irrespective of how much rainfall we receive!"  Thanks again, Don.

(chart 4-page-001.jpg)

(chart 3-page-001.jpg)

(chart 2-page-001.jpg)

(chart 1-page-001.jpg)

chart 4-page-001.jpg (43KB - 73 downloads)

chart 3-page-001.jpg (38KB - 79 downloads)

chart 2-page-001.jpg (54KB - 82 downloads)

chart 1-page-001.jpg (54KB - 82 downloads)