WINETASTER ON 11/01/04 WITH 7 JUDGES AND 9 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2004 Richard E. Quandt
FLIGHT 1: Number of Judges = 7 Number of Wines = 9 All the wines are Chateau de Beaucastel
Identification of the Wine: The judges' overall ranking:
Wine A is 1990 ........ 4th place Wine B is 1994 tied for 6th place Wine C is 1986 tied for 6th place Wine D is 1998 ........ 5th place Wine E is 1995 ........ 9th place Wine F is 1999 ........ 3rd place Wine G is 1993 ........ 8th place Wine H is 1991 ........ 2nd place Wine I is 2000 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H I Bob 1. 2. 4. 6. 8. 9. 7. 5. 3. Orley 6. 5. 4. 3. 7. 8. 9. 1. 2. Burt 7. 1. 8. 9. 5. 3. 6. 2. 4. Mike 4. 5. 6. 2. 8. 3. 7. 9. 1. Ed 7. 9. 6. 3. 2. 1. 4. 8. 5. Dwight 7. 9. 5. 4. 6. 2. 3. 1. 8. Dick 1. 4. 2. 7. 9. 6. 8. 3. 5.
Table of Votes Against Wine -> A B C D E F G H I
Group Ranking -> 4 6 6 5 9 3 8 2 1 Votes Against -> 33 35 35 34 45 32 44 29 28
( 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 = 0.0952
The probability that random chance could be responsible for this correlation is rather large, 0.7214. 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 Dick 0.2521 Orley 0.2427 Mike 0.1617 Bob -0.0921 Burt -0.3180 Dwight -0.5448 Ed -0.7563
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 I is 2000 2. ........ 2nd place Wine H is 1991 3. ........ 3rd place Wine F is 1999 4. ........ 4th place Wine A is 1990 5. ........ 5th place Wine D is 1998 6. tied for 6th place Wine C is 1986 7. tied for 6th place Wine B is 1994 8. ........ 8th place Wine G is 1993 9. ........ 9th place Wine E is 1995 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 5.3333. The probability that this could happen by chance is 0.7214 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 Bob Orley Burt Bob 1.000 0.450 0.017 Orley 0.450 1.000 0.067 Burt 0.017 0.067 1.000 Mike 0.217 0.133 -0.217 Ed -0.833 -0.483 -0.317 Dwight -0.667 -0.067 -0.050 Dick 0.767 0.433 0.033 Mike Ed Dwight Bob 0.217 -0.833 -0.667 Orley 0.133 -0.483 -0.067 Burt -0.217 -0.317 -0.050 Mike 1.000 0.233 -0.367 Ed 0.233 1.000 0.350 Dwight -0.367 0.350 1.000 Dick 0.050 -0.733 -0.183 Dick Bob 0.767 Orley 0.433 Burt 0.033 Mike 0.050 Ed -0.733 Dwight -0.183 Dick 1.000 Pairwise correlations in descending order 0.767 Bob and Dick Significantly positive 0.450 Bob and Orley Not significant 0.433 Orley and Dick Not significant 0.350 Ed and Dwight Not significant 0.233 Mike and Ed Not significant 0.217 Bob and Mike Not significant 0.133 Orley and Mike Not significant 0.067 Orley and Burt Not significant 0.050 Mike and Dick Not significant 0.033 Burt and Dick Not significant 0.017 Bob and Burt Not significant -0.050 Burt and Dwight Not significant -0.067 Orley and Dwight Not significant -0.183 Dwight and Dick Not significant -0.217 Burt and Mike Not significant -0.317 Burt and Ed Not significant -0.367 Mike and Dwight Not significant -0.483 Orley and Ed Not significant -0.667 Bob and Dwight Significantly negative -0.733 Ed and Dick Significantly negative -0.833 Bob and Ed Significantly negative
COMMENT: Four of the wines in this tasting also appeared in a tasting of Chateau de Beaucastel in April of 2002. In that tasting, the four wines were ranked 1991, 1994, 1990, 1995. In this tasting, they were ranked 1991, 1990, 1994, 1995. Clearly, among the wines ranked on both occasions, the 1991 is superior. In fact, it came out first in the earlier tasting and second in this tasting, suggesting it is an extraordinary wine. This is completely contrary to what most wine writers claim is the case for the 1991 vintage. This group generally preferred the younger over the older wines, which is contrary to conventional wisdom. The one exception was the outstanding 1991, which was first in our prior tasting and second in this one. It should be noted that Beaucastel goes to sleep for decades. Thus, the 1998, which is tasting very well today, may next be at this level in 2008.
Return to previous page