WINETASTER ON 01/06/03 WITH 7 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2003 Richard E. Quandt
FLIGHT 2: Number of Judges = 7 Number of Wines = 9
Identification of the Wine: The judges' overall ranking:
Wine A is Behrens and Hitchcock PS 1999 ........ 9th place Wine B is Stag's Leap PS 1998 ........ 7th place Wine C is David Bruce PS 2000 tied for 4th place Wine D is Rockland Petite Syrah 1996 ........ 8th place Wine E is Bogle PS 2000 tied for 2nd place Wine F is Rockland PS 1999 tied for 4th place Wine G is Turley Rattlesnake PS 1999 ........ 1st place Wine H is Stag's Leap PS 1999 tied for 2nd place Wine I is Field Stone PS 1998 ........ 6th place
The Judges's Rankings
Judge Wine -> A B C D E F G H I Frank 9. 8. 7. 5. 3. 6. 1. 2. 4. Grant 8. 9. 7. 6. 3. 5. 1. 4. 2. Burt 9. 6. 7. 8. 3. 2. 1. 4. 5. Bob 9. 5. 4. 2. 1. 8. 3. 6. 7. Orley 7. 4. 5. 6. 8. 3. 1. 2. 9. John 7. 1. 2. 8. 3. 4. 5. 6. 9. Dick 6. 7. 4. 9. 5. 8. 1. 2. 3.
Table of Votes Against Wine -> A B C D E F G H I
Group Ranking -> 9 7 4 8 2 4 1 2 6 Votes Against -> 55 40 36 44 26 36 13 26 39
( 7 is the best possible, 63 is the worst)
Here is a measure of the correlation in the preferences of the judges which ranges between 1.0 (perfect correlation) and 0.0 (no correlation):
W = 0.3980
The probability that random chance could be responsible for this correlation is quite small, 0.0044. Most analysts would say that unless this probability is less than 0.1, the judges' preferences are not strongly related. We now analyze how each taster's preferences are correlated with the group preference. A correlation of 1.0 means that the taster's preferences are a perfect predictor of the group's preferences. A 0.0 means no correlation, while a -1.0 means that the taster has the reverse ranking of the group. This is measured by the correlation R.
Correlation Between the Ranks of Each Person With the Average Ranking of Others
Name of Person Correlation R Burt 0.7459 Frank 0.7167 Grant 0.5356 Dick 0.4500 Bob 0.2929 Orley 0.2667 John -0.0667
The wines were preferred by the judges in the following order. When the preferences of the judges are strong enough to permit meaningful differentiation among the wines, they are separated by -------------------- and are judged to be significantly different.
1. ........ 1st place Wine G is Turley Rattlesnake PS 1999 --------------------------------------------------- 2. tied for 2nd place Wine H is Stag's Leap PS 1999 3. tied for 2nd place Wine E is Bogle PS 2000 4. tied for 4th place Wine F is Rockland PS 1999 5. tied for 4th place Wine C is David Bruce PS 2000 6. ........ 6th place Wine I is Field Stone PS 1998 7. ........ 7th place Wine B is Stag's Leap PS 1998 8. ........ 8th place Wine D is Rockland Petite Syrah 1996 --------------------------------------------------- 9. ........ 9th place Wine A is Behrens and Hitchcock PS 1999 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 22.2857. The probability that this could happen by chance is 0.0044
We now test whether the group ranking of wines is correlated with the prices of the wines. The rank correlation between them is -0.0840. At the 10% level of significance this would have to exceed the critical value of 0.4830 to be significant.
We now undertake a more detailed examination of the pair-wise rank correla- tions that exist between pairs of judges. First, we present a table in which you can find the correlation for any pair of judges, by finding one of the names in the left hand margin and the other name on top of a column. A second table arranges these correlations in descending order and marks which is significantly positive significantly negative, or not significant. This may allow you to find clusters of judges whose rankings were particularly similar or particularly dissimilar. Pairwise Rank Correlations Correlations must exceed in absolute value 0.70 for significance at the 0.05 level and must exceed 0.60 for significance at the 0.1 level Frank Grant Burt Frank 1.000 0.900 0.717 Grant 0.900 1.000 0.733 Burt 0.717 0.733 1.000 Bob 0.467 0.267 0.183 Orley 0.300 0.067 0.483 John -0.233 -0.367 0.217 Dick 0.633 0.633 0.433 Bob Orley John Frank 0.467 0.300 -0.233 Grant 0.267 0.067 -0.367 Burt 0.183 0.483 0.217 Bob 1.000 0.000 0.267 Orley 0.000 1.000 0.333 John 0.267 0.333 1.000 Dick 0.050 0.250 -0.083 Dick Frank 0.633 Grant 0.633 Burt 0.433 Bob 0.050 Orley 0.250 John -0.083 Dick 1.000 Pairwise correlations in descending order 0.900 Frank and Grant Significantly positive 0.733 Grant and Burt Significantly positive 0.717 Frank and Burt Significantly positive 0.633 Frank and Dick Significantly positive 0.633 Grant and Dick Significantly positive 0.483 Burt and Orley Not significant 0.467 Frank and Bob Not significant 0.433 Burt and Dick Not significant 0.333 Orley and John Not significant 0.300 Frank and Orley Not significant 0.267 Bob and John Not significant 0.267 Grant and Bob Not significant 0.250 Orley and Dick Not significant 0.217 Burt and John Not significant 0.183 Burt and Bob Not significant 0.067 Grant and Orley Not significant 0.050 Bob and Dick Not significant 0.000 Bob and Orley Not significant -0.083 John and Dick Not significant -0.233 Frank and John Not significant -0.367 Grant and John Not significant
COMMENT: This was a tasting os Petite Syrahs (PS) from California. Wine B was the Stag's Leap Wine Cellars, made by Warren Winiarski, while Wine H was the competitor, Stag's Leap Winery. These are wineries that are entirely different, but with similar names---and there has been much litigation over exactly this point.
Everyone agreed that the wines were very drinkable right now. In addition it should be noted that the alcohol levels in thesse wines were never less than 13% and ranged up to 14.5%.
The negative correlation of group prerences with price was particularly surprising, because the first place wine was the most expensive, and an order of magnitude more expensive than the second and third place wines. In other words, the first placed wine was ten times more expensive than the second and third place wines, and the last placed wine was five times more expensive than the second placed wine. The Bogle 2000 PS was simply an extraordinary wine selling at $10. The only other time that this has happened in this group was when a $15 South African Cabernet was judged superior to some equivalent first growth South Africans. The group thought that the wines were really delicious. PS is said to be the same as Durif; the question is, are there Durifs from France of the same quality?
Return to previous page