WINETASTER ON 04/16/01 WITH 8 JUDGES AND 7 WINES BASED ON RANKS, IDENT=N Copyright (c) 1995-2000 Richard E. Quandt
FLIGHT 1: Number of Judges = 8 Number of Wines = 7
Identification of the Wine: The judges' overall ranking:
Wine A is Ch. Pichon Lalande 1988 ........ 1st place Wine B is Ch. Sociando Mallet 1988 tied for 4th place Wine C is Ch. Talbot 1988 ........ 7th place Wine D is Ch. Gruaud-Larose 1988 ........ 6th place Wine E is Ch. Lynch-Bages 1988 tied for 4th place Wine F is Ch. Le Tertre Roteboeuf 1988 ........ 2nd place Wine G is Ch. Lafite 1988 ........ 3rd place
The Judges's Rankings
Judge Wine -> A B C D E F G Burt 1. 3. 4. 6. 7. 5. 2. Ed 3. 6. 7. 5. 1. 4. 2. John 1. 5. 2. 4. 3. 7. 6. Frank L. 3. 6. 7. 1. 4. 5. 2. Grant 2. 3. 7. 6. 5. 1. 4. Orley 5. 2. 4. 6. 3. 1. 7. Frank V. 2. 7. 6. 3. 5. 1. 4. Dick 1. 2. 7. 4. 6. 3. 5.
Table of Votes Against Wine -> A B C D E F G
Group Ranking -> 1 4 7 6 4 2 3 Votes Against -> 18 34 44 35 34 27 32
( 8 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 = 0.2132
The probability that random chance could be responsible for this correlation is rather large, 0.1152. 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 Grant 0.7388 Dick 0.5357 Frank V. 0.5000 Ed 0.1261 Frank L. 0.0360 Burt 0.0000 John -0.2342 Orley -0.3929
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 A is Ch. Pichon Lalande 1988 --------------------------------------------------- 2. ........ 2nd place Wine F is Ch. Le Tertre Roteboeuf 1988 3. ........ 3rd place Wine G is Ch. Lafite 1988 4. tied for 4th place Wine B is Ch. Sociando Mallet 1988 5. tied for 4th place Wine E is Ch. Lynch-Bages 1988 6. ........ 6th place Wine D is Ch. Gruaud-Larose 1988 --------------------------------------------------- 7. ........ 7th place Wine C is Ch. Talbot 1988 We now test whether the ranksums AS A WHOLE provide a significant ordering. The Friedman Chi-square value is 10.2321. The probability that this could happen by chance is 0.1152 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 0.79 for significance at the 0.05 level and must exceed 0.71 for significance at the 0.1 level Burt Ed John Burt 1.000 -0.071 0.143 Ed -0.071 1.000 -0.071 John 0.143 -0.071 1.000 Frank L. 0.000 0.536 -0.071 Grant 0.393 0.286 -0.393 Orley -0.321 -0.214 -0.250 Frank V. 0.036 0.357 -0.179 Dick 0.500 0.000 -0.071 Frank L. Grant Orley Burt 0.000 0.393 -0.321 Ed 0.536 0.286 -0.214 John -0.071 -0.393 -0.250 Frank L. 1.000 0.000 -0.714 Grant 0.000 1.000 0.429 Orley -0.714 0.429 1.000 Frank V. 0.500 0.536 -0.071 Dick 0.179 0.786 0.179 Frank V. Dick Burt 0.036 0.500 Ed 0.357 0.000 John -0.179 -0.071 Frank L. 0.500 0.179 Grant 0.536 0.786 Orley -0.071 0.179 Frank V. 1.000 0.393 Dick 0.393 1.000 Pairwise correlations in descending order 0.786 Grant and Dick Significantly positive 0.536 Ed and Frank L. Not significant 0.536 Grant and Frank V. Not significant 0.500 Burt and Dick Not significant 0.500 Frank L. and Frank V. Not significant 0.429 Grant and Orley Not significant 0.393 Burt and Grant Not significant 0.393 Frank V. and Dick Not significant 0.357 Ed and Frank V. Not significant 0.286 Ed and Grant Not significant 0.179 Frank L. and Dick Not significant 0.179 Orley and Dick Not significant 0.143 Burt and John Not significant 0.036 Burt and Frank V. Not significant 0.000 Burt and Frank L. Not significant 0.000 Frank L. and Grant Not significant 0.000 Ed and Dick Not significant -0.071 Ed and John Not significant -0.071 Orley and Frank V. Not significant -0.071 Burt and Ed Not significant -0.071 John and Dick Not significant -0.071 John and Frank L. Not significant -0.179 John and Frank V. Not significant -0.214 Ed and Orley Not significant -0.250 John and Orley Not significant -0.321 Burt and Orley Not significant -0.393 John and Grant Not significant -0.714 Frank L. and Orley Significantly negative
COMMENT: In general, these wines were very much alike, or as Inspector Clousot would have said, "these taste suspiciously like 1988 red Bordeaux wines." The big news was that we had Robert Parker's ratings to compare, and the most expensive wine that should have stuck out in the tastings was the Lafite, based on Parker's ratings, but we thought Pichon Lalande stuck out. However, Parker's tastings were done in 1991, which proves once again that it is impossible to tell what a wine will taste like ten years later.
Return to previous page