WINETASTER ON 04/04/05 WITH 7 JUDGES AND 8 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2005 Richard E. Quandt, V. 1.65
FLIGHT 1: Number of Judges = 7 Number of Wines = 8
Identification of the Wine: The judges' overall ranking:
Wine A is Hartenberg Shiraz 1995 ........ 4th place Wine B is Allesverloren Estate Cabernet 1990 ........ 7th place Wine C is Bertrams Cabernet 1987 ........ 8th place Wine D is Plaisir de Merle Cabernet 1995 ........ 3rd place Wine E is Nederburg Private Bin 103 1989 ........ 6th place Wine F is Le Bonheur Cabernet 1995 ........ 2nd place Wine G is Uitkyk Cabernet-Shiraz 1995 ........ 5th place Wine H is Thelema Cabernet 1995 ........ 1st place
The Judges's Rankings
Judge Wine -> A B C D E F G H Mike 7. 5. 8. 4. 6. 2. 3. 1. Ed 2. 3. 8. 1. 6. 7. 4. 5. Frank 6. 7. 8. 2. 5. 4. 3. 1. Bob 1. 4. 8. 7. 5. 2. 6. 3. John 5. 6. 8. 1. 4. 2. 7. 3. Dick 6. 7. 5. 8. 4. 1. 3. 2. Burt 4. 5. 8. 2. 6. 1. 7. 3.
Table of Votes Against Wine -> A B C D E F G H
Group Ranking -> 4 7 8 3 6 2 5 1 Votes Against -> 31 37 53 25 36 19 33 18
( 7 is the best possible, 56 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 = 0.4354
The probability that random chance could be responsible for this correlation is quite small, 0.0033. 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 0.7619 Mike 0.7470 John 0.7186 Frank 0.6707 Bob 0.4048 Ed 0.1437 Dick 0.1437
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 H is Thelema Cabernet 1995 2. ........ 2nd place Wine F is Le Bonheur Cabernet 1995 --------------------------------------------------- 3. ........ 3rd place Wine D is Plaisir de Merle Cabernet 1995 4. ........ 4th place Wine A is Hartenberg Shiraz 1995 5. ........ 5th place Wine G is Uitkyk Cabernet-Shiraz 1995 6. ........ 6th place Wine E is Nederburg Private Bin103 7. ........ 7th place Wine B is Allesverloren Estate Cabernet 1990 --------------------------------------------------- 8. ........ 8th place Wine C is Bertrams Cabernet 1987 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 21.3333. The probability that this could happen by chance is 0.0033 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 Mike Ed Frank Mike 1.000 0.048 0.833 Ed 0.048 1.000 0.286 Frank 0.833 0.286 1.000 Bob 0.286 0.143 0.095 John 0.548 0.286 0.667 Dick 0.571 -0.667 0.333 Burt 0.595 0.310 0.548 Bob John Dick Mike 0.286 0.548 0.571 Ed 0.143 0.286 -0.667 Frank 0.095 0.667 0.333 Bob 1.000 0.310 0.333 John 0.310 1.000 0.071 Dick 0.333 0.071 1.000 Burt 0.548 0.905 0.119 Burt Mike 0.595 Ed 0.310 Frank 0.548 Bob 0.548 John 0.905 Dick 0.119 Burt 1.000 Pairwise correlations in descending order 0.905 John and Burt Significantly positive 0.833 Mike and Frank Significantly positive 0.667 Frank and John Significantly positive 0.595 Mike and Burt Not significant 0.571 Mike and Dick Not significant 0.548 Mike and John Not significant 0.548 Frank and Burt Not significant 0.548 Bob and Burt Not significant 0.333 Bob and Dick Not significant 0.333 Frank and Dick Not significant 0.310 Ed and Burt Not significant 0.310 Bob and John Not significant 0.286 Ed and Frank Not significant 0.286 Ed and John Not significant 0.286 Mike and Bob Not significant 0.143 Ed and Bob Not significant 0.119 Dick and Burt Not significant 0.095 Frank and Bob Not significant 0.071 John and Dick Not significant 0.048 Mike and Ed Not significant -0.667 Ed and Dick Significantly negative
COMMENT: The wines drink as if they were more mature than their chronological age. One wine, the Bertram's Cabernet, was oxidized. The Hartenberg and the Allesverloren did not have much fruit. Le Bonheur, which was within one point of being top wine, was exactly the same as the winner of the 1999 tasting of South African wines. The tasting also shows that Thelema is a real powerhouse in South African wines. This tasting is very compatible with the 1999 tasting in that at the earlier tasting the top three winners were identical to the top three winners in today's tasting (although the relative rankings of the top three were not the same), which suggests a significant degree of intertemporal consistency in the group. Many of us in the tasting were prepared to buy substantial quantities of Le Bonheur. Reflecting on our report from 1999, which said that the wines are drinking noticeably well in spite of their youth, we can say that six years later they have matured well and are drinking nicely. Added on January 15, 2012: Comparing Report 14 with the present one, we note that five of the wines in the trwo tastings were identical. The rank correlation between the group rankings accorded to these five wines is 0.8. which is quite high and suggests a considerable amount of intertemporal consistency; a feature which we have failed to observe in numerous other tastings with replicated wines.
Return to previous page