But Ashenfelter's hypothesis is straightforward:
it is widely agreed that weather influences wine quality but a mere handful of facts about local weather determine a vintage.
His original equation follows:
Wine quality = -12.145 + 0.00117 *(Winter Rainfall)+ 0.0614*(average growing season temp) – 0.003868*(harvest rainfall)Weather-based vintage prediction is not a new idea. What is new is the notion that laptop computers can outperform the most sophisticated noses and palates.
Professor Ashenfelter added data for winter rainfall and then rigorously measured their statistical relationships to the most objective measure of quality he could devise: an index of auction prices for about 80 wines after they had matured in the bottle. His goodness of fit for actual auction data was quite good, especially for the Bordeaux and Burgundy regions of France.
Over the years, Prof. Ashenfeltere has refined the technique and adapted it, more or less, successfully. But I don't think the wine establishment is yet impressed.
