Research Rewind

April 27, 2018

By Maureen Moroney

Question: Does blending make a wine better?

It’s clear that the intricate art of blending by a skilled winemaker can elevate a wine. But at its absolute most basic, with no art involved, does the simple act of blending two wines together, in equal portions, create something better than leaving them separate?

In a classic study from 1962, published in the Journal of Food Science, Singleton and Ough attempted to find an answer. (“Complexity of Flavor and Blending of Wines,”

Their study’s abstract:

Thirty-four pairs of similar, commercially acceptable wines and a freshly prepared 50-50 blend of each pair were scored for quality by a small expert panel. The composite scores of the blends did not differ to even a low degree of significance from the score of the highest-scoring member of the pair. The scores of the blends were very highly significantly better than the scores for the low-scoring wines or the mean of the low- and high-scoring wine in each pair. In seven cases among the 34, the blend’s composite score was higher than that of the highest-scoring wine, and in no case was it lower than that of the lowest-scoring wine of the pair. Increased complexity of flavor is believed to be a major factor in the favorable effect of blending shown in this study; other possible effects are discussed.


  • Round one of sensory evaluation had the tasters give scores and comments on a set of 10 wines of a specific type (examples: dry red table wines, red dessert wines, dry white table wines) on one day, then a different type the next, and so on.
  • The average scores and comments were used to choose two wines from each day (and therefore of each type) to include in the 50-50 blend in the second round of sensory evaluation.
  • The two wines of each type were chosen based on having the most similar scores but the most contrasting comments.
  • The second round of sensory evaluation again presented each type of wine on different days, with the same 10 wines as previously, but also including a 50-50 blend of the two wines selected from each category.
  • All samples were coded and randomized, so that the wines were presented blindly as 11 similar but unrelated individual wines.


  • No blends received a lower average score than the lower-scoring wine of the pair.
  • Seven blends received a higher average score than either of their component wines.
  • The average scores of the blends were not at all significantly different than those of the higher-scoring components.
  • The average scores of the blends were very highly significantly better than the average of their individual components’ scores, and very highly significantly better than the lower score of the two.
  • Based on known sensory thresholds, the differences in scores between the blends and their individual components could not be explained by differences in sugar, alcohol, acid, or color.


  • “If the blending of two standard-quality wines seldom gives a product as bad as the poorer, usually gives one (at the 50-50 blend level) as good as the better one, and sometimes gives one better than either, blending … probably is of more value than previously suspected.”
  • “The evidence points to an increase in complexity as a major contributor to the quality increase. By this is implied the addition of flavors whose absence cannot be considered a serious deficiency but whose presence in the proper amount contributes favorably. A further implication is that a flavor that may be undesirable when recognizably strong, may be a contributor to complexity and therefore not undesirable if below the recognition threshold in the blend.”

The basic conclusion that the authors drew was that, generally speaking, an increase in complexity leads to an increased perception of quality. Now remember that this study was done with mostly arbitrary, 50-50 blends. And imagine what can be achieved through thoughtful, informed, judicious blending.

Questions to consider in your own winemaking:

  • What are some possible reasons for choosing NOT to blend?
  • When considering blend options, what is the goal that you have in mind?