WINETASTER ON 09/10/01 WITH 8 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2000 Richard E. Quandt
FLIGHT 1: Number of Judges = 8 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Margaux 1990 ........ 5th place Wine B is Ch. La Conseillante 1989 ........ 3rd place Wine C is Ch. La Conseillante 1990 ........ 1st place Wine D is Ch. Lafite Rothschild 1990 ........ 7th place Wine E is Ch. Lafite Rothschild 1989 ........ 8th place Wine F is Ch. Cheval Blanc 1990 ........ 4th place Wine G is Ch. Cheval Blanc 1989 ........ 6th place Wine H is Ch. Margaux 1989 ........ 2nd place
The Judges's Rankings
Judge Wine -> A B C D E F G H Hunt 7. 4. 1. 2. 3. 8. 6. 5. Burt 2. 5. 3. 7. 8. 4. 6. 1. Ed 8. 6. 5. 7. 3. 1. 2. 4. Orley 3. 2. 4. 5. 7. 1. 8. 6. Bob 1. 4. 5. 6. 8. 7. 3. 2. Frank 6. 2. 1. 8. 7. 4. 3. 5. Grant 2. 3. 1. 5. 7. 6. 8. 4. Dick 8. 7. 1. 2. 6. 3. 5. 4.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 5 3 1 7 8 4 6 2 Votes Against -> 37 33 21 42 49 34 41 31
( 8 is the best possible, 64 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 = 0.1838
The probability that random chance could be responsible for this correlation is rather large, 0.1726. 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 0.5524 Frank 0.5270 Grant 0.4762 Orley 0.2275 Dick 0.0952 Bob 0.0361 Hunt -0.0838 Ed -0.3593
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 C is Ch. La Conseillante 1990 --------------------------------------------------- 2. ........ 2nd place Wine H is Ch. Margaux 1989 3. ........ 3rd place Wine B is Ch. La Conseillante 1989 4. ........ 4th place Wine F is Ch. Cheval Blanc 1990 5. ........ 5th place Wine A is Ch. Margaux 1990 6. ........ 6th place Wine G is Ch. Cheval Blanc 1989 7. ........ 7th place Wine D is Ch. Lafite Rothschild 1990 --------------------------------------------------- 8. ........ 8th place Wine E is Ch. Lafite Rothschild 1989 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 10.2917. The probability that this could happen by chance is 0.1726 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 0.74 for significance at the 0.05 level and must exceed 0.64 for significance at the 0.1 level Hunt Burt Ed Hunt 1.000 -0.333 -0.333 Burt -0.333 1.000 -0.190 Ed -0.333 -0.190 1.000 Orley -0.286 0.357 -0.214 Bob -0.333 0.690 -0.429 Frank 0.024 0.333 0.238 Grant 0.286 0.643 -0.690 Dick 0.452 0.000 0.238 Orley Bob Frank Hunt -0.286 -0.333 0.024 Burt 0.357 0.690 0.333 Ed -0.214 -0.429 0.238 Orley 1.000 -0.048 0.262 Bob -0.048 1.000 0.190 Frank 0.262 0.190 1.000 Grant 0.524 0.405 0.333 Dick -0.024 -0.405 0.143 Grant Dick Hunt 0.286 0.452 Burt 0.643 0.000 Ed -0.690 0.238 Orley 0.524 -0.024 Bob 0.405 -0.405 Frank 0.333 0.143 Grant 1.000 0.048 Dick 0.048 1.000 Pairwise correlations in descending order 0.690 Burt and Bob Significantly positive 0.643 Burt and Grant Not significant 0.524 Orley and Grant Not significant 0.452 Hunt and Dick Not significant 0.405 Bob and Grant Not significant 0.357 Burt and Orley Not significant 0.333 Frank and Grant Not significant 0.333 Burt and Frank Not significant 0.286 Hunt and Grant Not significant 0.262 Orley and Frank Not significant 0.238 Ed and Dick Not significant 0.238 Ed and Frank Not significant 0.190 Bob and Frank Not significant 0.143 Frank and Dick Not significant 0.048 Grant and Dick Not significant 0.024 Hunt and Frank Not significant 0.000 Burt and Dick Not significant -0.024 Orley and Dick Not significant -0.048 Orley and Bob Not significant -0.190 Burt and Ed Not significant -0.214 Ed and Orley Not significant -0.286 Hunt and Orley Not significant -0.333 Hunt and Burt Not significant -0.333 Hunt and Bob Not significant -0.333 Hunt and Ed Not significant -0.405 Bob and Dick Not significant -0.429 Ed and Bob Not significant -0.690 Ed and Grant Significantly negative
COMMENT: These were all extraordinary wines and it was surprising that for young wines they had unusually little tannin. One participant felt that the wines are drinking beautifully but have substantial tannins which are soft and sweet. In the four comparisons of vintages for the same chateaux, in three cases the 1990 was preferred. The first place wine was the 1990 La Conseillante and the least preferred wine was the 1989 Lafite. The 1990 Margaux was in fifth place and was a 100 point Parker rated wine in 1993. Based on the across the board performance for the two vintages, La Conseillante was the winner and Lafite was last.
Return to previous page